MITLibraries

MITLibraries Document Services Room 14-0551 77 Massachusetts Avenue Cambridge, MA 02139 Ph: 617.253.5668 Fax: 617.253.1690 Email: docs@mit.edu http://libraries.mit.edu/docs DISCLAIMER OF QUALITY Due to the condition of the original material, there are unavoidable flaws in this reproduction. We have made every effort possible to provide you with the best copy available. If you are dissatisfied with this product and find it unusable, please contact Document Services as soon as possible. Thank you. Pages are missing from the original document. PAGE 46 IS MISSING FROM ORIGINAL A REEVALUATION OF THE VALUE LINE INVESTMENT STRATEGY; CAN ACTIVE COMMON STOCK PORTFOLIO MANAGEMENT PRODUCE SUPERIOR RETURNS? by JOHN HARRY FEINGOLD 91 B.S., Massachusetts Institute of Technology (1978) SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF IASTER OF SCIENCE at the MASSACHUSETTS INSTITUTE OF TECHNOLOGY (JANUARY 19, 1979) Copyright John H. Ferngold Signature of Author. De Certified by . - . tment of Manag ¥- ent, (.-.............. January 19, 1979 J' Thesis Supervisor _ -: ?:7 . -, Accepted by................................................. Chairman, DeXAtmeEt "iM 5 , .";' :- OF *'iC:i'O LO Y LIBRARIES of Management Committee . i j :a .' -2- A REEVALUATION OF THE VALUE LINE INVESTMENT STRATEGY; CAN ACTIVE COMMON STOCK PORTFOLIO MANAGEMENT PRODUCE SUPERIOR RETURNS? by JOHN HARRY FEINGOLD Submitted to the Sloan School of Management on January 19, 1979 in partial fulfillment of the requirements for the degree of Master of Science ABSTRACT Value Line Investment Survey asserts that their Group 1 ranked securities have outperformed lower ranked stocks with remarkable consistency for over twelve years. However, many have criticized their tests as inconclusive. This thesis is a Value Line performance analysis covering the period of November 1971 through December 1977. The sample includes the 100 stocks ranked 1 for year ahead appreciation, assembled in an equally weighted portfolio. The test utilizes regression analysis to compare the excess returns on the Value Line portfolio (Value Line portfolio returns -Treasury Bill returns), the dependent variable, and the excess market return, (three different indices, equally and value weighted, Standard and Poor's "500")the independent variable. The Value Line ex-post alpha, for trading on the publication date (-5, 0, 5, 10, 20 days delay analyzed), regressed on the equally weighted portfolio (the best performing of the market indices), indicates a positive 12% yearly extra return, with a t-statistic of 3.6 , significant at over the 99% confidence level. -3- This result, significant and large, suggests that Value Line recommended investment strategies which consistently outperform the market, contradicting extensive literature documenting the efficiency of capital markets. Thesis Supervisor: Fischer Black Title: Professor of Finance -4TABLE OF CONTENTS Abstract.................. 4 ......... ..................... Table of Contents......... Table of Tables.......... Acknowledgement 2 ....... .. ......................... .. ......................... ....... 6 .................................. 7 ........... Preface................... .................................. ................................. 10 Introduction .............. Chapter 1 Value Line: Chapter 9 What Does t All Mean? .................. .... 11 2 Previous Research on Value Line Investment Survey........ 25 Chapter 3 Financial Theory - Efficient Markets - Random Walk .......31 Chapter 4 Methodology .......... 3... ..........................36 41 Computer Program 1 Analysis ..................... Computer Program 2 Analysis ....................41 42 Computer Program 3 Analysis ..................... 47 Computer Program 4 Analysis ..................... 48 Computer Program 5 Analysis ..................... Chapter 5 Results .................. . ... . .. . .. .. .. .... . .. ... .. .. .. .51 Chapter 6 Taxation and Trading Strategy........................... 78 -5- Chapter 7 Conclusion, Summary and Topics for Future Research .......93 Appendix 1 Cumulative Company List ... .. .... Sample Portfolio ............ Appendix .. 97 ............. 115 ................ 2 Computer Programs ....... ............................... 116 FORTRAN Program 1. ............................... 118 FORTRAN Program 2. ............................... 121 FORTRAN Program 3. ............................... 126 FORTRAN Program 4. ............................... 130 TSP Program 5 .... ............................... 131 Appendix 3 Regression Output....................................... 132 -5 Days Delay.......... .......................... 133 0 Days Delay ........ .......................... 136 5 Days Delay.......... .......................... 139 10 Days Delay.......... .......................... 142 20 Days Delay......... .......................... 145 Glossary of Terms and Equations. .......................... 148 Footnotes ....................... Bibliography..... .......................... 150 .......................... 155 ............. --6TABLE OF TABLES Table 1 Record of Value Line Rankings for Timeliness Allowing for Changes in Rank ...................... 18 Table 2 Record of Value Line Rankings for Timeliness ......19 Table 3 Percent Change in Price - Value Line Equally ......21 Weighted Portfolio.......................... Table 4 Total Return - Equally Weighted Value Line ............ .. 22 Portfolio........................... Table 5 Total Portfolio Growth - 1971-1977........ Graph 1 Total Portfolio Growth ................... Table 6 ket Summary of 1971-1977 Data-Value Line vs. Mar ke~t Indices ..52 , eJJ ...... .60 ......................... Graph 2 Extra Return 1971-1977 Data vs. Delay in Day s ..... 61 Table 7 Stock Purchase -5 Days Trading Delay........ ...... 65 Table 8 Stock Purchase on Day of Value Line Receipt. Table 9 Stock Purchase 5 Days Trading Delay......... ...... .67 ...... 66 Table 10 Stock Purchase 10 Days Trading Delay........ ..... Table 11 Stock Purchase 20 Days Trading Delay........ ...... .69 Table 12 Treasury Bill Returns 1971-1977....... ...... .71 Table 13 Cumulative Market Returns - Zero Days Delay. ...... .73 Graph 3 Cumulative Portfolio Growth........... Table 14 Yearly Growth Portfolio............... ..... · 68 .......74 ...... .75 Table 15 Approximate Composition of Value Line Portfo lio... 85 -7ACKNOWLEDGEMENT The author wishes to express his deep gratitude to Fischer Black, Professor of Finance at the Sloan School of Management, for suggesting this project and for guiding it through to its conclusion. The author is also indebted to Professor Irwin Tepper of the Harvard Business School for his patient and clear explanations, unfailing support and friendship over the past year. The author further wishes to express his appreciation to Heather and Jean Titilah for their invaluable manuscript suggestions, and painstaking editing and typing. The author gratefully acknowledges John Maglio and Charlene Mahoney of the East Campus Computer Facility staff for their advice and programming assistance. I also appreciate the guidance and interest of the entire Sloan School Finance Department, particularly Professors Robert Merton, Carliss Baldwin, Donald Lessard, John Cox and Dan Holland, and also the Professor Himself, Arnold Barnett. This work was made palatable by the frequent support and assistance of my friends through each stage of the project. Special thanks to, Ann Short, Jennifer Katz, Larry Gordon, Steve Miller, Rob Goldenhill, Stan Brenner, Brian Ison, Harriet Schwartz and Miriam Sherburne. -8- Finally, the author wishes to express his wholehearted love and appreciation to his parents for their continuous guidance, encouragement and support throughout not only this project, but this author's entire life. -9- PREFACE The author encourages comments, suggestions and questions regarding this thesis. dence Please direct correspon- to: John H. Feingold 535 East 86th Street New York, New York 10028 (212) 879-7982 (212) 249-5117 or my current address will be on file at the Sloan School of Management's Alumni Office: Sloan School Alumni Office E52-460 Sloan School of Management Massachusetts Institute of Technology 50 Memorial Drive Cambridge, Massachusetts (617) 253-7168 02139 -10- INTRODUCTION Samuelson, in a short article,1 touches upon many of the points which the careful reader should keep in mind while reading this performance evaluation of the Value Line investment service. He speaks of the advantage of portfolio diversification, and warns that holding a large number of securities does not in itself ensure efficient diversification since they may be strongly positively correlated. He also speaks of portfolio performance evaluation, and the appearance of superior performance by increasing risk through leverage in upmarket periods. Use of regression analysis and the capital asset pricing model will aid in a clearer understanding of the results of this study. Samuelson also warns that even if an investment advisory service such as Value Line should beat the market, it may be due only to chance. Within this thesis the statistical tests will be performed to provide proof or rejection of the null hypothesis at a statistically significant level. After reading this paper, one question is still sure to remain. If Value Line has produced better results than a passive strategy, does this imply anything for their expected future performance? -11Chapter VALUE LINE: 1 WHAT DOES IT ALL MEAN? Arnold Bernhard, Research Director of the Value Line Investment Survey, in a 1970 presentation at the University of Chicago, presented exhibits which indicated Value Line Investment Survey had a consistently good predictive ability. He said that unless his results of statistical analysis are pure luck, they contradict the random walk hypothesis. Fischer Black, my thesis advisor, entered this debate at the conference by presenting the evidence for passive portfolio management. He felt that the Value Line performance results were impressive, but that the statistical tests were inadequate. His objection is based on the fact that Value Line order ranked utilizing cross-sectional tests which indicated little about statistical significance. Black prefers the use of regression testing for consistency of performance. It is this method which I will employ in the original research contained in this thesis. The objective of an actively managed portfolio of common stocks is to choose securities so that there is a greater return than when an index fund is purchased. The choice of misvalued securities must overcome the consequential costs of churning the portfolio. In addition, a -12- market timing approach can be taken to take advantage of financial market movements, and hopefully help avoid having funds invested during declining periods in the equities markets. The active/passive investment decision will be examined in detail, including an overview of current research in financial theory of efficient markets. Value Line ranks 100 stocks in Group 1, the highest category for year ahead performance. Implied is an advantage to holding more than one security, otherwise investing a large portion of your assets in what they consider the number one best stock would be recommended. 3 Harry Markowitz analyzed investor behavior and discovered that investors generally try to maximize returns while avoiding risk. He believes that the important charac- teristics of a portfolio of stocks are the expected return and the riskiness. Intelligent rational investors should naturally hold that combination of risky assets which maximize expected returns for a given degree of risk. Markowitz identified that with the knowledge of a securities expected return, variance, and covariance with the market, that efficient portfolios could be created. This is the theory which Value Line simplifies into the recommendation that each investor should hold at least 16 to 25 of their Group 1 securities, and also the reason I chose to analyze the portfolio of all Group 1 securities. The general formula for -13- computing the variance of a portfolio is: 2 N N i=l i=l up= x.x i y coviy xi = proportion invested in security i xy = proportion invested in security y coviy = covariance between the rates of return on i and y also CAB rab S A SB where CAB = covariance between return on A and return on B rab = coefficient of correlation between return on A and return on B SA = standard deviation of A SB = standard deviation of B These equations show that diversification does not help when security returns are perfectly positively correlated. However diversification can eliminate risk with perfectly negatively correlated securities. In the case of partial correlation as found in Value Line portfolios Group 1, diversification lends an advantage, since in theory an investor is only rewarded for bearing market risk represented -14- by beta rather than total risk6 represented by sigma - or what Value Line has computed by the name "Rank for Safety." Value Line recommends that if one holds fewer than 15 of their Group 1 securities, the "safety rank" or total risk view, is a valid measure. 15 securities, in different But if one holds more than industries, one should be only concerned with betas, since the portfolio would then be fairly well diversified. In order to better understand how Value Line makes stock recommendations, let us examine the criteria used in their performance studies and analyses. Value Line computes the rank of approximately 1700 stocks each week. The financial analysis is condensed into two numbers, one conveying information about how the stock rates in the sample of about 1700, relative to expected price performance in the next 12 months. The second measure is an indication of investment safety, or total risk of the individual security. Value Line publishers explain their rankings as follows: Value Line rank definitions, 100 stocks Rank 1: Expect the best price performance relative to the other stocks covered in the survey. 300 stocks Rank 2: Expect above average price performance. -15- 900 stocks Rank 3: Expect price performance in line with the market. 300 stocks Rank 4: Expect below average price performance. 100 stocks Rank 5: Expect poorest price performance in relation to the other covered stocks. Other studies of Value Line have examined the stocks in each of these rank portfolios results hold true. to determine if the expected Value Line claims "not every stock will perform in accordance with its rank in every year. But such a high percentage have in the past for logical reasons based on earnings, growth rates, and risk, that the probabilities definately stand in your favor when you line up your stocks with the Value Line Ranks." If Value Line does have the ability to discriminate between securities in the ranking method, an obvious strategy would involve the purchase of Rank 1 securities and the short sale of Rank 5 securities. literature Although according to Value Line this would be a strategy, it is not examined in the paper. I studied only those securities ranked 1, to determine whether their Rank 1 performance was superior to that of various market indices. Value Line is mailed on a schedule that is aimed to assure delivery to the subscriber on Friday. In my analysis I refer to "x" days delay in acting upon Value Line advice. A zero delay means you would -16act on the Friday you receive the survey. A five day delay is five Stock Exchange days, or most likely seven days. Like- wise, a delay of 20 Stock Exchange days would be about four weeks. I also analyze a negative delay, a hypothetical strategy of acting on the advice five days early. I consider this important both academically and because it is a strategy which could be ollowed by a person who understands how the Value Line ranks are computed. In a letter of June 2, 1978 written by Samuel Eisenstadt, Chief Statistician of Value Line to Fischer Black; "Most subscribers receive the survey on Friday, some even on Thursday. The rankings are determined 7-12 days prior to the subscriber's receipt of the survey ... subscribers that are acquainted with the mechanics of the ranking system can successfully anticipate rank changes by following earnings reports in the Wall Street Journal. For example if a Group 1 stock comes out with a poor quarterly earnings report he need not wait 7-12 days to be told that the ranking has been lowered." Therefore I thought it would be interesting to look at the five day negative time delay. rankings appear to be about a week However although 'stale' by the time the subscriber receives the survey, I am sure that in the case of a major development, the ranking could be altered until the survey is printed on Wednesday evening. I do not believe in the likelihood that a subscriber could duplicate the results obtained by following Value Line a week early. -17- However the zero day delay is the strategy that subscribers could duplicate. Value Line's performance record is regularly reported in their publication as demonstrated in Tables 1 and 2. These two tables, assembled by Value Line, show the results that an investor would have received following Value Line recommendations from April 1965 through July 5, 1978. The Value Line analysis assumes, in the case of no allowance for rank changes, that an investor buys an equal dollar amount of each stock of each rank at the start of each year and holds an unchanged portfolio for the entire year. the next year the portfolio is rebalanced. At the start of Allowance for rank changes in the portfolio is updated weekly. There are no allowances for transaction costs. The compilation of Value Line table statistics utilizes geometric averages of price changes in each period. When dealing with portfolio performance,9 a compounding of arithmetic averages of price changes would have simulated an actual portfolio strategy. Tables 3 and 4 present data compiled by Value Line applied to an institutional universe of common stocks -Standard and Poor's "500" stock Composite Index. This analysis was executed to disprove the belief by some that Value Line is capable of discriminating only in that segment of the market made up of small, inactively traded "secondary -18- o\o No UN L. H + c:n o kCo N + + + + r + - 1o -- Ln - o\o r -H H H I' t, in 5 LIn) a (N I I I I 1 )>{ . Q4 Q) r- ' H- o a) Co H-- 0. H(N k~ mcn co M M I N I Cr '.) I I o\O r * - + + H * + I 0 * H- H C Ln H Q i H H M w CD) 0.) D 0)a -H H r- c_Q 1-4 F- ZH2 H + + o 0 + + I o C C' m QQ >~ N rro ,-4 O W QC H>U Ln 0 + Co rI ra r:4r 0 .3 E-1b LO Ln N- I I I O N 0 ' 0 o\O CO N (N t o H + + o\o t I C fU) Co H o\O ) N. (N + I Co MC m H - N o I I I I N ~4 Lfl 0i HN r-q o\O U r-I H C + Ln * z O HIn r-I r * m 0.) H > H M rs H0)O 0 ~\0 co r o00 o o\O z co VN N O C- 0 0 C co I 0 zH 0O_ M + r- I o\° I'*D + N + i + O (N I I I I o Co Co O H (N + + Hl 00 (N N + 00 n + H + L.) N + N H + -M H CN + rCo N) H-q o\o - o N I) I I IC LI I H o\O o +n OD co + o C4 (N 0 0 0 0 (N I 0) rH M O 0 C + + + H : CC O' Q N' I' r) H kO N H + + + + + o\O 0 (N H + I H o (N (N I + o\o r I' + r-l H o I 0' H N 0m . H (N I d: 0 0 LO 0 0 Co + Q4 > k0 H-l L) + H H C + 0 (N + + + (N M t LO S24 04 aQ 04 Q. O O O O O 0 0 0 0 0 m0 N + 4 0 > -19o\o O Ln aH O 0' M N + + Ln + Ltn r4 H ( 00 N H + + + £O 0 'r Ln aCO I + C Ln N (N cc C rn r3 rn 03 I I I ! I o\O r H d' >1 H ul) 00 N- 0. (N I N C . n co CO I N I ) (N (N o' Ln . (N I. + H (N 04 + I + H 0 z N C (N Cl W 0 H Ino\o + + Z < O Iu o\O I I Lo 0 + + H u H H I I I 4 + +,1 z00 a0 co H - H E:Z o\° r-- .{ . r. 0 co -I : O 0 o' a) Hi o\O (N + H o\O z r-Ij r-4 + o\o w u H W9 r 00 .. D a) tl) L "C o C- O H1 I CI (N I C.' I (N I a. H H v a. H + O 0 + U (N + o. (1N + C C N N I I I H q O c 00 f< Ez 0 + 5: 0H . I H I" Z (N I H z H H (N I N EL 0H N I o\o N H zH 4 Ln o\o H Ei H l + + + + a -K o\o 00c O LE} H + H . (N + 1 a) 1.- N I H o\O ' H I o\O oL d C) 0-.' m.'- O CO LI aO ' + . + N H + N (N cc cc LO I (N It H Ccc C coo c. N H + + o\o rn -K O ' ° Iu + N ° N - H + + I + o\O c 00 + H CN O :3:3 O O V3 V3 V 00 (oN H + cH o 0n?3 I LO : ::::3 O -H + r- 1 O0 C U) . O) - < 0 H C?. co C?' In C. + (n o. + m + m + r--i N CY) "IT . In . + + ,--t ,3 ¢ O O 0 L9 0 ~4 v: L) L U2 - CI O O O 0 0L9 L4g> - L9 0 O U0 -20- stocks" that may be inefficiently priced. Therefore Value Line assembled this data to demonstrate performance on that segment of the market considered by most to be the most well analyzed and efficiently priced. The rankings assume that a position was taken at the beginning of each year and held for 12 months without ranks changes. 1 0 Table 3 summarizes recent changes in price for an equally weighted portfolio. Table 4 contains total returns figures, change in price plus dividends, and is also equally weighted. These tables show that results for the Standard and Poor's "500" stocks show discrimination, and question the validity of efficient market theory. Value Line describes their statistical analytical technique as "Investing in Common Stocks." I will summarize Value Line's criteria for computing a rank for price performance of the next 12 months. Their four main criteria are: 1. Non-parametric value position 2. Magnitude of over or underevaluation 3. Earnings momentum 4. Earnings surprise factor The non-parametric value position of each stock concerns a price-earnings measure. Relative earning and prices of all Value Line stocks for the same period. price momentum 11 factor is also included in order to help A -21Ln . ,-. "~) -. co Ln co , JCN o -- I Lr) la: I- r-l z iC) (N (N I I I (N IV) N4°° H NO0l O o H ,a) o\o I C? ' On I'l 0 O -IJ c 0 L* co o * 0 0 0 4 S N (N + Ln -,- N -H + m) O c H + 0O N C?) + + a, 3) Hl -I r-- nd rH · k4 0 . I (N O C I.) C O cN CC H -- I O I 00 0 c p-.f H3 3. O Hl N U'3 ,-i H 0 0 w c Lr) rl: Z: -3 In) I O N H I 00 1V (N I I O CO n ICN I I C N a 0c Ha) 0-P o' H t-- 24 H 0 0 z0Pl) cN I z C)1 4 I z U r-- H L) a( HC? N N N r tO cN o CN < N co r.J1 * :4 E-A H-I K I * O LC N C O Ln H 0 00 m 00 NN C cc I N H H I 0 H I CI · o . oo --I CN Ql H 0 r.. 4 O k QJ O r. C) I0 r .o (N N (N O (N cc IC) C?) dr m 0 I 4 :: O rO 0 a4 I H ( 0 CN N CN ccC?) H : (N 0 m -O Ic : .C L tH r In m I oc 0 0 03 0 0 0 0 0 0 0 Oc ) ~- - -22oo ,cr LnoO Ln 0 rHl LO iC. IO) O 0 H H rl- rn Ln o .(N fl: H Q 4 3 O H N H zW Eq a) 0 LCD d I z C O n) Ln o N (N O'(N H./ In 0c 'I o; o ,Q o o '.o (* r- 0 0 0 ,o o c rn cc urs to o- Ln . 4 N · ri H HI~ cc 0 O~ N (N c--l z U f:: 4 N zz H rH + + zo D r4 a) 0 I- * rdc 4 O m?) C I cc I O O rC H -qH I cN I (N l n Lr (N CN N (N H 0 (N C) C? £C ) * H CM * N fn co * * Cy) ? n ~A E- *k co Ns O IC N- * rm O LC H H H N- co 0 Ln D O N N H- I I N I Ln H .--I %D l r CO cn (H o\o r-l H H H O H C) 0 0E- N r- ,C co H r- O -- l N H N N C~~~~~ 'N II N z E4 H W r r-- rJ co C ,c o C) - rl o 0O 0 k 0k O r,. r,.O LC I C? I H co 0o H 0O 0 r..D 0O oo - 1 ko H -Q= -- or cn l C I 0 I 0 .-- o' ILn co ko 0 (n O C?) H r N U ( r co :: 0 m I M 0O 0['..40 C Or r Cl 6 H C? 04(a) 0 0 CO o0 0 lo * 0 u0 k k 09 03 V @ -23- predict future action. earnings, This measure is used to combine rank, price rank, and price momentum into one figure. The magnitude of over or underevaluation is a measure used by Value Line's analysts to "measure the disparity between current price-earnings ratio of a stock and its historical norm. l2 Earnings momentum is a function of the year-to-year change in quarterly earnings per share of each stock. The earning surprise factor seems to be a convenient way to integrate new or unexpected information into the ratings. This explanation is what Value Line claims they do to estimate stock values. Of course we don't know their method for sure -- however this is what they claim to do. Utilizing these many measures of performance and analytical predictive ability, Value Line attempts to refute the efficient market hypothesis by demonstrating accurate active portfolio management. Burton G. Malkiel defines the random walk as "The history of stock-price movements contains no useful information that will enable an investor consistently to outperform or buy and hold strategy in managing a portfolio." 1 3 Scholars have defined three forms of the efficient market hypothesis. The weak form asserts that current stock -24prices reflect all information from historical prices, i.e., one cannot apply a mechanical formula to past stock prices to beat the market. The semistrong form of the hypothesis is that in analyzing public knowledge or charting the underlying companies does not produce superior investment results. The strong form does not produce superior investment results. The strong form goes so far as to say that even those individuals with insider information cannot make use of this information to produce superior stock market returns. 1 4 these strong forms of the efficient market hypothesis are true, it appears Value Line could not achieve superior stock market returns. If -25Chapter 2 PREVIOUS RESEARCH ON VALUE LINE INVESTMENT SURVEY A survey of recent financial literature on previous empirical research on Value Line, produced five outstanding performance analysis articles. This section summarizes the findings and relates the highlights of earlier works. John P. Shelton in "The Value Line Contest - A Test of the Predictability of Stock Price Changes, "1 studies the results of a large number of individual investment decisions on the 1965-1966 Value Line contest. motional device. The contest was a pro- Appropriately named, a contest of stock market judgment, which attracted 18,565 entrants. Each contestant chose 25 stocks from a list supplied by Value Line made up of securities they ranked four or five on November 26, 1965. The rules assumed that each contestant would form an equally weighted portfolio with a value computed at the close of the market on December 31, 1965. Value Line analysts selected their own portfolio of stocks from those that were ranked Group 1. Prizes were awarded to those individuals who chose portfolios which outperformed Value Line by the greatest margins. The question Shelton asks is "Did the 18,565 contestants select portfolios, on average, that differed from the performance of the 350 stocks to which the selection was -26- confined by enough to conclude that the result was not likely to have happened by chance?" 2 Shelton found that the 350 stocks ranked four and five experienced a 5.9 value over the 26 week contest period. loss in The range of stock price changes for individual securities ranged from a doubling in value to 202.5% to a two-thirds decline in value to 64.7%. Shelton discovered that the average score achieved by the contestants was approximately 49 standard deviations greater than the expected mean. He states, "It is extremely unlikely that a difference as large as this would have occurred if the price changes during those six months were so truly random." 3 Shelton seriously questions the existence of efficient markets, based on the superior performance of a large sample of individual investors. This doubt is limited, of course, to the six month period of the Value Line contest. Warren H. Hausman in "A Note on the Value Line Contest: A Test of the Predictability of Stock-Price Changes" calls attention to what he considers an inappropriate statistical test used in Shelton's paper. He concludes "The fact that investors (or contest entrants) tend to agree with each other (as Shelton found) need not mean that they know anything of value. Neither does the fact that, on a single occasion, they outperformed a random selection of stocks."4 Hausman suggests additional observations during different time periods. -27- John Michael Murphy studied and performed analysis on the 1969 Value Line contest. Murphy's results are in agreement with Shelton's and cast doubt on the usefulness of the random walk hypothesis when describing the stock market. The 1969 contest rules differed from those in 1965. Con- testants were allowed to choose their portfolios from any of the 1,258 stocks Value Line analyzed at the time. Murphy's conclusions is: One of "The results reported are significant and inconsistent with the spirit of the random walk hypothesis, but statistical, logical, and methodological considerations preclude a claim that the hypothesis has been rej ected."6 The next Value Line inquiry was conducted by Robert S. Kaplan and Roman L. Weil in an article, "Risk and the Value Line Contest. " 7 This contest covered the period from August 18, 1972 to February 16, 1973. The authors hypothe- sized that given efficient markets, stock prices would simultaneously adjust at the publication or release of all new information, including analysis performed by Value Line. They believed that a high beta, high risk, portfolio should do well when the market rises, and perform poorly when there is a general drop in price levels. A low risk portfolio should perform better than the high risk portfolio when there is a market decline. -28Kaplan and Weil did not believe they could pick 25 stocks that could outperform the market. They thought that it made sense to enter two portfolios, a very high beta portfolio, and a very low beta portfolio. direction, If the market moved in either they could take advantage of the situation. The high beta portfolio had a beta of 2.13 and the low risk portfolio by their estimates a beta of 0.21. The average Value Line rank of the two portfolios was about equal, which would naturally lead to an expectation from Value Line of about equal performance. Their results were computed during a period when prices for all stocks in the Value Line survey declined 6.7%. The author's low beta portfolio actually increased in value 3.8%, placing it in the top 2,043 of the 85,744 portfolios entered. The high beta portfolio declined 22.9% and scored in the bottom 519 of the 89,744 portfolios entered. The results support the authors' hypothesis. They showed the expected results according to the beta theory of portfolio performance. In their conclusion, Kaplan and Weil say "The rankings are flawed, since much of the variation in performance is caused by differences in the risk of stocks in each group." 8 The Kaplan and Weil study chose portfolios with equal Value Line rank but with different levels of risk. The return on these portfolios differs by more than 26 percentage points. During the same time period, Value -29Line Group 1 and Group 5 portfolios had approximately equal risk in a beta context, but return differed by only 3 percentage points in performance, despite differential Value Line ratings. They concluded that detailed investigation of individual securities is not worthwhile and that stock market movements are dominated by systematic risk. The next study of Value Line was performed by Professor Fischer Black, "Yes Virginia, There Is Hope; Tests of the Value Line Ranking System." 9 This thesis is closely modeled after Black's paper, with three important differences. 1) Black's observations of Value Line rankings and stock prices were monthly, while mine are weekly; 2) Black looked This study at all the Value Line ranks in his study period. looks into Group 1 rank and compares Group 1 stock performance to various market indices; 3) Black did his study in cooperation with Value Line, who performed the actual computations. This thesis is independently written and researched. Black states "According to the analysis that Value Line performed with my help, its ranking system appears to be one of the few exceptions to the rule that attempts to separate good stocks from bad stocks is futile."1 0 Professor Black observed of Value Line, "The system tends to assign high marks to stocks with low price earning ratios, relative to historical norms, and relative to the price earnings ratio of the market." 1 1 -30- Discussing shortcomings of previous Value Line analysis, Black says "Cross-sectional tests generally tell you nothing about statistical significance, or whether performance in one period is likely to be repeated in future periods." 1 2 Black favors regression testing which is the method he suggested for this thesis. He also examines the implications of transaction costs and he discusses ways of utilizing the ranks to minimize transaction costs. In concluding, he states "The net result of the portfolio simulation, assuming transaction costs of two percent or less in and out, was that the strategy continued to give significant results over the five year period, although the level of significance was reduced somewhat." 13 This has been a survey of the research leading up to my analysis of Value Line. Each study concentrated on a slightly different aspect of active portfolio management. Nevertheless each concluded that there is at least some question as to the validity of the efficient market hypothesis. Hopefully this performance study will provide another piece of evidence. -313 Chapter FINANCIAL THEORY - EFFICIENT MARKET - RANDOM WALK The efficient market hypothesis is at the heart of this research. If Value Line can discriminate undervalued securities, then surely a re-evaluation is due to the presupposed degree of efficiency in markets. This section will provide an overview of the most well known previous investi- gations into efficient markets and stock market performance studies. In a test of efficient markets, Fama 1 studied the proportionate price changes of the 30 Dow Jones industrial stock for the period 1957-1967. very low, the average being 03.2 He found serial correlations To adjust for the domina- t:ion in his correlation coefficients of just a few extreme observations, Fama also examined only signs, (+ or -), rather than size of successive days, statistics to determine if runs persisted. He found a negligible departure from randomness. HHis study provides strong support for the random walk hypothesis.3 Another test of efficient markets was performed by William F. Sharpe.4 1963. He studied 34 mutual funds from 1954 to Sharpe takes account of rates of return as measured by total risk, variability, pricing model. 5 utilizing the capital asset The capital asset pricing model expresses a linear relationship between risk and return on a portfolio. -32- This presupposes the condition that the portfolio is efficient, meaning that the portfolio provides maximum returns for a given level of risk. Sharpe found that if expenses of the funds are ignored, 15 of the 34 funds outperform the Dow Jones industrial average after risk adjusting. If expenses are considered only 11 funds beat the Dow Jones average while 23 did worse. Sharpe's conclusion is that mutual funds failed to consistently outperform the market, implying an efficient He states "... support(ing) to the view that capital market.6 market is largely efficient and that good managers concentrate on evaluating risk and providing diversification, spending little effort (and money) on the search for incorrectly priced securities."7 Another study of mutual fund performance was under- taken by Michael Jensen8 covering the performance of 115 mutual funds from 1945-1964. Jensen understood that variations in fund performance was expected because of difference in the fund's risk. individual randomly His method involved comparison of an fund's performance with the performance of a selected portfolio of equal risk. 9 When ignoring mutual fund expenses, about half of the funds did better, and half worse than expected. "This is the result that would be expected if the market were highly efficient with market prices fully reflecting all that was knowable through public announcement or -33ascertainable through the efforts of individual security analysts. 10 When accounting for expenses only 43 out of 115 mutual funds showed superior performance. Jensen's work in the evaluation of mutual fund performance provided additional support for the random walk hypothesis. "The evidence on mutual fund performance discussed above indicates not only that these 115 mutual funds were on average not able to predict security prices well enough to outperform or buy and hold policy, but also that there is very little evidence that any individual fund was able to do significantly better than that which we expected from mere random chance."ll William Sharpe in "Adjusting for Risk in Portfolio Performance Measurement" 12 shows that adjusting for risk properly is the way to accurate performance evaluations. S.harpe describes methodology used in this Value Line study involving the computation of the excess return for each period, being the difference behind a portfolio return and Treasury Bill returns.13 He states that in times of greatly varying short term interest rates, it is essential to utilize the excess return methodology to obtain meaningful study results. He also-speaks of a reward to variability, measure and a model of naive investor behavior as necessary parts of a performance study. Exactly what he is advising is to be aware of risk, as it is important measure on performance -34evaluation. William Sharpe wrote another article "Likely Gains from Market Timing"l 4 which I include in this overview because of the suspicion that Value Line may attempt market timing decisions in the adjustment of the average beta of their portfolios. This article tries to address the question, how superior must one's predictions be to implement a market timing style effectively? "Attempts to time the market are not likely to produce incremental returns of more than four per cent per year over the long run. Moreover, unless a manager can predict whether the market will be good or bad each year with considerable accuracy (e.g., be right at least seven times out of ten) he should probably avoid attempts to time the market altogether." 1 5 Jeffrey Jaffee in "Special Information and Insider Trading" 1 6 provides insights regarding the strong form of the efficient market hypothesis. He summized that only "intensive trading samples yield profits greater than commissions" regarding usefulness of insider information.17 study also suggested a profit opportunity: than is found in this Value Line study. Jaffee's but one smaller Roger Ibbotson also supports the efficient market hypothesis as it applies to new issues of common stock. "We cannot reject the hypothesis that an investor in a single random issue has an equal chance for a gain or loss ... The results generally confirm -35that there are no departures from market efficiency in the after market. 18 Fischer Black and Myron Scholes lend insight into the determinant of active portfolio management. They state that as an individual trading on superior information, such as Value Line recommendations, you are competing with other individuals who are following the same strategy: "As a group those who trade on information cannot make money. If some individuals make money by deviating from the market portfolio, then other individuals must lose money by deviation from the market portfolio. All individuals together hold the market portfolio... An investor who trades on information incurs substantial costs from his activity. He may spend money gathering and analyzing the information he uses. He incurs transaction costs when he buys and sells. He may realize gains that he does not have to realize, and then pay taxes sooner than he has to. He holds a portfolio that is not as well diversified as the market." 19 Black and Scholes state that although information traders do not generally earn superior returns, they help keep the market efficient by integrating all information into current stock prices. Black contends that the cost of in and out brokerage changes on a $40.00 stock is about 3% for a round lot, but this was before negotiated commissions. This research all seems to say that Value Line must show impressive performance to overcome all of the disadvantages of active portfolio management and provide its subscribers. with risk adjusted superior returns. -36- Chapter 4 METHODOLOGY This section of the thesis describes topics ranging from data collection to regression testing. The intent is to provide a precise explanation of the research in enough detail to support further research and expansion by a future historian. It is important for the reader to understand that this performance study was done independently of Value Line, or of any other commericial enterprise. No direct contact was made with the Value Line organization, nor was any data supplied directly by them. All Value Line information was hand collected from publicly available sources. The data from November 5, 1971 to December 30, 1977 was collected and put into machine readable form. The Value Line Group 1 portfolio, each containing 100 stocks, were recorded on a weekly basis. The Dewey Library at M.I.T., Harvard's Baker Library, Boston Public Library as well as the libraries at Northeastern, Boston University and Wellesley College, were cooperative in :Locatingback dated Value Line reports. Mr. Evan Shulman of Batterymarch helped fill in the final gaps. Study was concentrated on those 100 stocks Value Line ranked highest for year ahead performance. Initially we thought this would encompass hand coding the 100 stocks -37each week for the entire 321 week period. It is more efficient to code only the initial portfolio on November 5, 1971, and record only the weekly additions and deletions. With this information a computer program can generate weekly stock portfolios, resulting in a 90% savings in information quantity which it was necessary for computer readable form. This method provided ease in obtaining many spot accuracy checks. Each week's portfolio updates were recorded on 3x5 index cards which were headed with the date. The analysis is limited to those Value Line securities listed on the New York and American Stock Exchanges. This is because the CRSP data base included only those securities. However, the index card weekly update file contains listings of overthe-counter securities for possible future research efforts. Upon completion of the index card weekly update file, each company name was manually coded with its unique identifier, an eight digit ICUSIP number. It is this identifier, supplied by the Stock Exchange, which is punched onto weekly update computer cards to act as input to the portfolio generation program. a list containing The ICUSIP numbers were coded from all stocks on the above mentioned exchanges. Whenever a stock name from the index card file was not located on the computer generated tape identifier number list, it was assumed to be a security listed on the over-the- -38counter market and therefore outside the realm of this study. Roughly three per cent of the Value Line Group 1 recommendations are over-the-counter securities. I believe ignoring these stocks will have little effect on this analysis since I see no obvious reason that Value Line should be better able to discriminate over-the-counter securities. One may consider the over-the-counter portion of the market less efficient, and more open to fundamental and technical analysis. If that were truly the case, Value Line would include more over-the-counter securities in their universe. Compilation of the portfolio update list is straightforward for the two most recent years. For this period, Value Line presents newly added securities to Group 1 rank with a box next to the name. From 1975 chronologically reversed, the task is more difficult in that weekly updates are not easily identified. The method is standardized by reliance on the "Summary of Advice Section," a list of 100 Group 1 rank for year ahead performance. Adjacent week lists are compared, and additions and differences were discerned by discrepancies in the adjacent lists. The Value Line publication is only satisfactory in transmitting their analyst's recommendations ber. to the subscri- The publisher lets pass many spelling and alphabetiza- tion errors serious enough to cause doubt in the subscriber's mind as to exactly what action was recommended. -39Value Line utilizes an up arrow symbol (A) to indicate an upward valuation in rank, similarly a down arrow indicates a drop in rank (V). Mysteriously, in copies of the reports I found triangle arrows pointing in most conceivable rotations. The recorded errors in Value Line are diverse. In some cases, the same stock was dropped from Group 1 in two consecutive weeks, or just claim to drop it, only to confuse the subscriber by it's appearance as a Group 1 on the following week. Although the restraint of the portfolio remaining stable at the 100 stock level implies an equal number of additions and deletions per week, this is not always the case. Prior to 1975, the Value Line information supplied was more difficult to utilize. For example, on April 11, 1975 the recommendation was made to drop Norton-Simon from Rank 1. This was confusing considering that the previous week's Group 1 did not contain Norton-Simon. Value Line also recommended dropping Outlet Company, which was not in Rank 1. On July 21, 1974 Value Line identified Tyler Corporation as in Group 2 with an up arrow (A), signifying that it had just been promoted from Group 3. However, in the Summary of Advice section, Value Line recommended that this stock be dropped from the Group 1 portfolio. The arrow had been reversed. Furthermore, the July 21, 1974 survey may have been incorrectly dated July 19, 1974. Value Line recommended -40adding Entex to Group 1 on April 29, 1977, even though it was already in the Group 1 portfolio. On December 28, 1973 Barber Oil was listed with an (A) as an addition to Group 1, however it was not added to the Summary of Advice Group 1 list. To compound the error, and maintain 100 stocks in the portfolio, Value Line continued to maintain Greater Washington Investments as Group 1, while listing it as a drop candidate with a down arrow was dropped (V). Again on December 28, 1973, Koehring from the Summary of Advice's Group 1 -- however, :Ltwas not labeled with a 2 down arrow (), but was maintained a Group 1. It was lowered in rank a week later, without mention of the apparent discrepancy. A more serious error occurred on December 21, 1973 when 101 stocks were listed as belonging to Group 1. In trying to cope with these errors one must rely on the Summary of Advice section as being correct and one must behave as a typical, rational investor. Note that one cannot go back to the original reports for verification since those full page reports are published only four times a year, and therefore are likely to be out- dated. When the data was coded it consisted of 917 different companies with roughly five additions and five deletions each week. Some weeks there were no changes to the portfolio, and other weeks, those which coincided with many quarterly company earnings reports publication contained -41roughly 30 portfolio updates. The data span a 321 week period. Computer Program 1: Weekly Portfolio Generation The function of computer Program 1, which was written in FORTRAN, was to translate the initial stock portfolio and weekly updates into a sequence of data identified weekly portfolios. For input, the Program required the initial portfolio and data cards in the following format: Date + or - YYMMDD + ICUSIP Number (8 digit) Stock 1, Stock 2, Stock 3 .... The Program generated output to a disk storage file. This output consisted of: weekly portfolios, cumulative stock list and number, date list and week number. As a verification of sample data check, the program detected and printed error messages if the add or drop symbol was incorrect, if an attempt was made to drop a company not in the current portfolio, or if one tried to add a stock that already existed in the Group 1 domain. tained updated portfolios in numerical The program main(alphabetical) order. The programs are reproduced in the Appendices. Computer Program 2: Stock Name Portfolio Verification Program 2 was designed solely for data verification purposes. It consists of two parts. The first reads the CRSP (Chicago Research in Security Prices) computer tapes and stores the names of the companies previously identified -42by ICUSIP numbers. The second half of the Program prints weekly portfolios, identifying companies by name. Inputs included two volume CRSP data tapes, disk file of weekly portfolios from Program 1, and a disk file of the cumulative company list from Program 1. The output is a cumulative list of companies names, a printout of weekly stock portfolios by name, and a list of bad ICUSIP numbers, that is, numbers not found on the CRSP tapes. This may be due to read errors, or more likely because there is no company issued to that ICUSIP number, in which case a keypunch error is uncovered. This output was compared to the original Value Line survey to provide data verification. Computer Program 3: Variable Vector Displacement Returns Generation. Program 3 performed all of the computation for the thesis. It required two megabytes of core storage and ten cylinders of temporary disk storage. (Including tape mount- ing charges it cost roughly $50.00 per run.) Inputs to Program 3 included complete CRSP daily stock returns, 2 volume tapes, a cumulative list of 917 companies, a weekly data list, weekly portfolio composition, and calendar of vector addresses of returns on tape. Output of Program 3 contains a disk file of weekly market returns, a disk file of weekly Group 1 portfolio returns, and cumulative returns. -43- This program computed weekly market returns from the daily returns on the tape. It also calculated Group 1 portfolio weekly returns from the individual stock daily returns and then formed the 100 stock portfolio. Two decisions were made when writing Program 3. The first was to equally weight the Value Line portfolio of Group 1 securities. The second involved the choice of the market indices to read from the CRSP tape. This program also provided the researcher with the option to simulate different purchasing acquisitions delays around the Value Line publication date. It was run for each of these variations: Days (-) Acceleration/Delay Before Trading on Value Line Recommendation Run 1 -5 Business days, prior Friday Run 2 0 Delay, publication date Run 3 5 Business days, delay, subsequent Friday Run 4 10 Business days, delay, two weeks Run 5 20 Business days, delay, four weeks Most investors receive their copy of the Value Line survey in the Friday mail. (This was discovered by examination of the library reception stamps on survey copies.) Due to the Friday market closing, those prices were used as the base, zero delay case for my analysis. -44When deciding the choice of a market index, some believe the best choice was one that most closely represented the whole market. Such a decision is independent of this, or any other survey. Another view involves comparing Value Line's performance with a passive strategy demonstrating a high correlation, mixed with lending or borrowing. could be a technical buy and hold strategy. This The comparison with a highly correlated index would provide a good measure of whether Value Line does really do better. Definition of a surrogate for the market is difficult. Therefore, one must compare the Value Line portfolio to the three indices; the equally weighted market, the value weighted market, and the Standard and Poor's "500" composite. The Value Line portfolio is assembled weekly, buying an equal dollar amount of each stock ranked 1, selling at the end of the week, producing a return figure and repeating weekly. The weighting of an index reflects the representative importance of each stock. Value weighted indexes are dominated by the larger high capitalization weighted companies. Equally indexes give greater weight to smaller companies. A value weighted index oriented or ranked strategy can be followed by all investors when the individual stocks are weighted in relation to the company's value. In order for all investors to hold an equally weighted portfolio, major capital redistribution must occur. -45- /J ' ;. The Standard and Poor's "500" composite index includes 425 industries, 25 railroads and 50 utilities. The market value of the stocks in this value weighted index comprise about 80% of the value of the New York Stock Exchange. Professor Fischer Black made the suggestion that an equally weighted index of the stocks listed on the New York and American Stock Exchanges, may behave similarly to a value weighted proxy of all capital assets, which is really the ideal index representation from some points of view. Another important issue regards market indices and portfolio evaluation as methods of averaging. These two most commonly used methods are an arithmetic mean, or a geometric mean. Value Line uses geometric averages for their in-house research. Most others use arithmetic measures. Indices based on geometric means will increase more slowly and decrease more rapidly than an index based on an arithmetic mean. Utilization of an arithmetic mean makes more sense for comparative performance research. It corresponds to the performance that could be duplicated by an investor who rebalances his portfolio each period, to include equal dollar weights of each included stock. N t-h The geometric mean is the root of the product of N observations. mean is the simple average. The arithmetic PAGES (S) MISSING FROM ORIGINAL PAGE 46 IS MISSING -47- Computer Program 4: Ibbotson - Sinqufield Treasury Bill Program The only program run in the batch mode was the one to read the Ibbotson - Sinqufield tapes. These tapes provide the monthly U.S. Treasury Bill returns. This index of short term rates, of Treasury Bills of the shortest maturity greater than one month, was utilized as the risk free rate in the capital asset pricing mode equation. The tapes M.I.T.'s Sloan School of Management own (supplied by the Center for Research in Security Prices) end in 1976. I coded by hand the 1977 returns according to the Ibbotson - Sinqufield procedure: RFT = PF, T/Pr, (T-l) -1 Where T=month, R=Treasury Bill Returns, Pr=Price. I collected the data from the Wall Street Journal. All monthly returns were compounded to a weekly series for the regression. The program environment consisted of FORTRAN, executed on the conversational monitor system operated under the IBM Virtual Machine Facility 370. The interactive mode of operation made data correction, location and verification easier than under a batch system. The CRSP computer tapes provided daily stock returns for the American Stock Exchange universe. and New York Stock return is defined as a change in total value of an investment for a common stock -48for a period. The returns are completely adjusted for dividends and other distributions. The last program was written in TSP, Time Series Processor. It calculates ordinary least square linear regressions on the return data generated by the previous programs. The output of TSP included estimates of regressional coefficients, estimates of standard errors, and t-statistics for the null hypothesis that individual regression coefficients are not zero, and R TSP produces a least square linear regression equasion such that: y = a + bx such that Z (y, -y) ( z is minimized -x) (y, -y) (x 1 -x) a = y - bx coefficient of determination r 2 -2 = Z (y, -y) Z (y, -y) t-statistic (a) or (X) (y, -2 N-2 -49- The TSP program takes as inputs; weekly portfolio returns, weekly equally weighted market returns, weekly Standard and Poor's "500" returns, and weekly Treasury Bill returns. The program generates excess returns variables and performs the following regressions: Dependent Variable Independent Variable (Value Line Portfolio (Equally weighted port- -Treasury Bill Returns) folio (Value Line Portfolio -Treasury Bill) (Value weighted portfolio -Treasury Bill) - Treasury Bill) (Value Line Portfolio --Treasury Bill) (Standard and Poor's port/ folio Regressions are performed -Treasury Bill) for the total period as well as yearly sub-periods. Professor Black recommended the regression method of testing which he utilized in his Value Line survey because of its ability to demonstrate consistency of performance. In order to compare different portfolios a method must be used to relate the beta, or relative risk, and interpret the rate of return. Adjustment for the effect of beta were conducted by examination of the extra return of the Value Line portfolio after regressing the excess returns of Value Line on the excess returns of the market (by subtracting out the Treasury Bill Rate). The extra return is represented by the alpha (a) in the regression equasion. The Value Line portfolio excess return is the dependent variable. -50- RVL -RF = a p (Rm RF)+ p VL - weekly return on Value Line portfolio RF = one week Treasury Bill rate, updated monthly R = return on marked index m VL = beta of Value Line aVL = extra return of Value Line portfolio EVL = error term, assuming normal distribution should be zero If the market is truly efficient, and Value Line does not have the ability to discriminate over/under valued securities, alpha (a) should be zero. The t-statistic of the alpha is the important statistic, dividing alpha by the standard error, a t-statistic greater than 2 is considered to allow rejection of the hypothesis that alpha is zero at the 95% confidence level. In estimating alpha, considered the measure of Value Line's performance, the effect of varying risk is adjusted regardless of general economic conditions. Market movements should have no effect on this measure of performanc e. -51Chapter 5 RESULTS This thesis tests the hypothesis; the Value Line investment strategy produces no extra return on a market portfolio over an eight year test period. Regression testing adjusts for different sensitivity and risk of various portfolios. presented. In addition, simple growth figures are These are not risk normalized, but are still a good performance measure because the test period is an almost flat period for the value weighted portfolios in the growth of equities. market indices. This is verified by the performance of Observations were weekly unless identified otherwise. Table 5 summarizes the results of unadjusted market growth over the 1971-1977 period. During this eight year period an initial investment in the Standard and Poor's "500" index would have just maintained its dollar value, (non-inflation adjusted), growing to only 1.01 times its initial investment. (This is an arithmetic average of the returns presented on the Table.) A value weighted index did slightly better growing to 1.28 the initial investment. The equally weighted portfolio showed better performance growing to 2.35 times the initial investment. It is difficult to say which index best describes the true market movements for the period. I favor the -52- r.)1o :q r-I WdO > > o Co '" ~ O' IF Un I'll m CN N ON Om o c o 0 0 0 co o0.~ Q. o d 0 Ln 0 C r-l d 0O J 0O o30~ II 0 r'- rlH N- rrN LO , LIr-I oO * r 0 3) r--q r--I r- * k 0 * 3 cUGc 0 > Ed {- -Q :3: U) 0 .E 0' 0 .H 0 0 0E~ 0 ra) O 4- 0 a >1 .--I ,I cd N N LL 0 (N ( CN N .) ::3 d W E-q 3.) o d -H o ,, a, m :> d d ! LO' 0 r-H 0 N -53Graph 1 TOTAL PORTFOLIO GROWTH Value Line = V.L. Equally weighted = E.W. Value weighted = V.W. Standard and Poor's "500" = S&P Growth 6 4 3 JL EW 2 VW -- 1 -5 0 5 10 Days to act on information S&P 20 -54- equally weighted index. It represents a broad selection of construction of the Value Line portfolio presented in this paper. This point may be inconsequential however, since, as mentioned earlier, the choice of a market index may be independent of my Value Line portfolio construction. It appears that smaller companies performed better than the large capitalization companies for the survey period. This fact helps explain the superiority of performance of the equally weighted index. In evaluating the growth of the Value Line portfolio, one may notice that if one pursued the hypothetical but impossible strategy of purchasing the recommended stocks five days before receipt of the recommendations in the mail, one would have realized growth of 5.8 times the initial investment. tor. This would be impossible for the casual inves- However, it is possible to keep track of current earning reports, and be especially observant of falling earnings -- leading to an early prediction of an upcoming Value Line drop recommendation. It is more difficult to predict a Value Line upcoming add recommendation. The investor must scan the set of 300 stocks ranked 2 to search for sharply increasing earnings which may signal an upgrading in rank. Whereas the investor must only scan the 100 Group 1 securities in search of a drop recommendation. -55Furthermore, academicians may find this result fascinating, since it is very likely after the date on which the Value Line computer oriented mechanized stock valuation formula is trading date on which Value Line utilizes the applied a information to manage their supervisory funds. To purchase stocks on the same day as receipt of the publication is the first strategy that can easily be adhered to. Following would grow 4.8 times. this strategy, the initial investment These figures make no provision for transaction costs, brokerage commission, or taxes which would become due when following a trading strategy. Those taxes would be deferred when following a buy and hold strategy such as buying one of the indices. Even with these negative factors, 4.8 times growth compared with 2.3 times growth for the equally weighted portfolio has value as a strategy and could absorb considerable expenses and continue to maintain its superiority. Investment strategies simulating delays of 5, 10, and 20 days before purchase or sale of the Value Line Rank 1 securities were tested. times respectively. Growth was 3.2,. 2.9 times and 2.5 Notice growth was perfectly inversely ranked with trading delay. If one followed the alternative of delaying a month before acting upon the Value Line recommendation, the total portfolio growth of 2.5 times would be very close to the growth of the equally -56- weighted portfolio of 2.4 times. Of course the active trading strategy is subject to many expenses the buy and hold strategy avoids. Even waiting four weeks after the report publication, the Value Line return does not drop below that of the best of the market indices. In addition, it was observed, non risk adjusted, that for all trading delays, in terms of total growth, Value Line did better than the value weighted portfolio and the Standard and Poor's "500" index. During this same period, an investment in Treasury Bills grew to 1.4 times the initial investment. Treasury Bills are considered the risk free investment. The reader should notice that the risk free investment outperformed two of the market indices, doing better than the value weighted combined New York and American Stock Exchange index, and the Standard and Poor's "500" index. It is not wise to rely heavily on the unadjusted growth figures without considering the regression results. The Value Line portfolios are not as well diversified as the market indices and not all of the portfolio movements are described by the market. When stocks were purchased 2 on the Friday of publication, R terms were: R 2 .85 for regression on the equally weighted portfolios; R2=.79 for regression on the value weighted portfolios; R =.73 for regression on the Standard and Poor's "500". Correlation -57- coefficients are: Equally weighted portfolio .92 Value weighted portfolio .89 Standard and Poor's "500" .86 This demonstrates that most of the Value Line portfolio movement is explained by market movement. It is this divergence from returns predicted by market movements that allows the portfolio to achieve superior performance. If the Value Line portfolio is a riskier portfolio, modern portfolio theory would lead us to expect extra return to compensate for this risk. Modern portfolio theory states that an investor should be rewarded only for the non-diver- sifiable risk associated with the market. There should be no benefit for holding company specific risk. One would expect an extra reward to induce holding an undiversified stock portfolio. is riskier, In addition, if the Value Line portfolio (more volatile), than the market indexes, in an up market, it would be expected to produce more growth. This growth however would not be a measure of special predictive ability, but the result of holding a risky portfolio. Of course, it is possible that there exists a market timing element in the Value Line analysis which indicates the proper periods in which to shift into high or low beta portfolios. Naturally, Value Line should choose -58- high beta portfolios when an up market is anticipated, and low beta portfolios if a downturn is evident. If a downturn is evident, however, another strategy is to get out of stocks completely, or invest in negative beta securities. This question in regard to Value Line must be studied elsewhere. Table 5 also provides the answer of a valid question. Suppose Value Line really does not have any ability to discriminate. Suppose that the subscribers have been led to believe that the Value Line analysts have predictive ability. This may just be a rationalization for spending the approximately $3100.00 for the yearly Value Line subscription, or it may be they like executing trades with their brokers, enjoy receiving mail, reading the Value Line research, or are non-value maximizers. If investors automatically followed Value Line advice, the recommended stocks would rise in price in the short run. However, if there was really no information content in the Value Line report, one might suspect that by the time one month elapsed, the fairly efficient market would once again correctly price each security . The growth figure for Value Line with the 20 day delay would be below that of the equally weighted portfolio. Therefore, Value Line recommendations probably contain some information not yet disseminated to the efficient market before publication. -59Table 6 will allow us to quantify the Value Line advantage. It summarizes the results of the 1971-1977 regressions, together with the important statistics, the alpha, t-statistic for the alpha, the beta coefficients calculated for each trading delay purchase of sale. All of the R 2 were high and the t-statistics of the betas were very high (10 and above)). The striking observation derived from Table 2 is the perfect rank correlation of alphas for each of the market indices. Notice all of the alphas are positive. In all cases, they decrease with increased trading delay. The alphas are adjusted to be meaningful yearly percentage indicators. I believe the most profound single figure of this thesis is presented in Table 6. Taking market actionon the Friday of publication, the Value Line portfolio, when regressed against an equally weighted market index, (the best representative proxy for the market), resulted in a positive alpha of 12. significant beta of 1.04. with a t-statistic of 3.6 and a This strongly suggests that Value Line recommendations do produce investment strategies which consistently outperform the market by over 10%. figure is large enough to absorb considerable expense. This transaction -60- (a H 2 rr-- I 0) r-- Cd rl- PP ; ; 0 z H w o3 U) a4 Cd a) rQ * * 0o Id 1l 14 1l 1 r4 0 .H r-l 0 4 .H r-f O 44 -P1 40) 0' a) -P Xi 4o 02 En I -W ~4 .,H 4i l co LC 0 Ln rl- ~I L 1 rl 4 -, M 0\0 I 4J 'P0 d Q4 . Vn 0 En 40 03 . ri HH rl Q 40 02 Cd Ln O m - 0 0 0 0 40 · ro 4n 0 43 Cd m rd ~4 fa q3 m rd . 0 11 0\0 O * 0 0 Cd H H Cd r-I r-4 co mI 0 P i·0' HH r000 r) O C) O r- 0 CD .H U4 o\ p0 , 40 .rl 0 Hi * CD U H z o0 0 0 0 HI ~ C L ; 4 CNC r r r o\o d H o 'H .rq 4O cn 0 Q) c,.,H Cd 4o >1 >1 >1 >>1 rdrdrdrd r 40 0a > 0 Cd4 u Q) k4 iL O Ln oo ,-- rdrdrrrd Ln 0L C) CN I rl IINo ro o o H--H LO CD O (1 -61Graph 2 EXTRA RETURN 1971-1977 DATA VS. DELAY IN DAYS Value Weighted = V.W. Equally Weighted = E.W. Standard and Poor's "500" = S&P alpha 30 25 20 .S&P 15 VW 10 5 EW -5 0 5 TRADING DELAY 10 20 -62- Concentrating on the equally weighted portfolio, the alphas were positive and statistically significant for -5 days and 0 days delay, 15. (t-statistic 4.5) and 12. (t-statistic 3.5) respectively. The time delay between acting on the Value Line recommendation waiting a week are the most noticeable returns. immediately and in terms of extra After a week's delay the alpha is reduced to 4.5 (t-statistic 1.4) with a low t-statistic which does not disprove the null hypothesis at the 95% confidence level. For delays of 10 and 20 days the alphas were lower: 3.2 and 1.6, with low t-statistics of 1.0 and .49. The beta of the regressions varied from 1.04 to .99, all approximately equal to the market beta of 1.0. Of course all of the market indices do not have the same level of risk. This would suggest that an equally weighted index, accentuating the effect of smaller companies compared to a value weighted index, stressing larger companies, would be more volatile, since the smaller companies in general may be riskier than the larger ones. Both the value weighted and the Standard and Poor's "500" index show the same rank ordering of alpha over the entire study period. However, from Table 6, recorded alphas are higher for 0 days delay; 22.2% (5.66) for value weight and 26.2% (5.95) for the Standard and Poor's "500". The Value Line portfolio regressed against these -63- two market indices produced significant t-statistics, (over 2), for all trading day displacements. In the same way as demonstrated earlier the tstatistic declines as the trading delay increases. Alpha decline - 5 to 20 days Trading Equally weighted 15. to Value weight 26. to 12. Standard & Poor's "500" 30. to 16. 1.5 This table confirms that regardless of which market index one prefers, there is a Value Line strategy that produces large significantly positive alphas, ignoring transaction costs and taxes. Tables 7 through 11 provide yearly extra return figures, t-statistic and betas. Market Rundown Equally weighted Value weighted 5.% 17.% Down year -26.% -17.% 1974 Down year -20.% -22.% 1975 Up year - 58.% 31.% 1976 Up year 49.% 18.% 1977 mixed 17.% -2.% 1972 Up year 1973 -64- Table 7 summarizes results for -5 trading days for the three market indices. ranging from 38.% to 1.7%. All of the alphas are positive However, for many of the yearly periods, the t-statistic does not demonstrate an alpha appreciably different from zero at the 95% confidence level. However the majority of our conclusion can be drawn from the total period results. I would not look for significance within the yearly figures. Table 8 presents yearly data for 0 days market action delay. This table shows two negative alphas. Both are less than 1% yearly and they have very low t-statistics. The equally weighted portfolio regression demonstrates alphas of 23.% and 24.% for 1973 and 1974 with significant t-statistics. The value weighted indices show significance for 1974 through 1977 with yearly alphas in the 30% range. Table 9 represents yearly data, observing a five day delay before acting on Value Line recommendations. The general level of t-statistics decreased with four negative alpha observations. Otherwise, in general, alphas are positive and large. Table 10 summarizes results for a two week, day), trading delay period. (10 Once again there are four negative observations of alphas. Although a t-statistic of -1.6 on a two tailed test is not significant at the 95% -65Table 7 SUMMARY OF DATA FOR STOCK PURCHASE 5 DAYS BEFORE VALUE LINE PUBLICATION Equally Weighted Portfolio Year Alpha% t-statistic Beta 1971-1977 15. 4.5 1.0 1972 1973 1974 1975 1976 1977 10. 24. 28. 12. 4. 7. 1.7 3.2 3.3 .90 .90 .52 .86 1.0 1.1 1.0 1.0 1.1 Value Weighted Portfolio Year Alpha% 1971-1977 1972 1973 1974 1975 1976 1977 26. 1.6 1i. 33. 31. 22. 39. t-statistic 6.5 .36 1.1 2.3 3.1 2.9 6.1 Beta 1.1 1.1 1.2 .89 1.3 1.5 1.3 Standard and Poor's "500" Year 1971-1977 1972 1973 1974 1975 1976 1977 Alpha% t-statistic Beta 30. 6.4 1.1 4.0 9.7 34. 37. 32. 47. .77 .82 2.2 3.3 3.7 6.4 1.1 1.2 .85 1.3 1.4 1.2 -66- Table 8 SUMMARY OF DATA FOR STOCK PURCHASE ON DAY OF VALUE LINE RECEIPT Equally Weighted Portfolio Year Alpha% 1971-1977 1972 1973 1974 1975 1976 1977 12. 10. 23. 24. 6. -0.92 4.1 t-statistic 3.6 1.7 3.4 2.8 .48 -.10 .92 Beta 1.0 .89 .99 1.1 1.1 1.1 1.5 Value Weighted Portfolio Year Alpha% 1971-1977 1972 1973 1974 1975 1976 1977 22. -0.34 15. 27. 28. 21. 31. t-statistic 5.7 -.74 1.3 2.1 2.7 2.6 5.4 Beta 1.1 1.1 1.2 .88 1.3 1.5 1.3 Standard and Poor's "500" Year Alpha% 1971-1977 1972 1973 1974 1975 1976 1977 26. 1.4 14. 28. 35. 31. 40. t-statistic 5.9 .28 1.1 1.9 3.0 3.5 5.9 Beta 1.1 1.0 1.2 .84 1.3 1.4 1.2 -67Table 9 SUMMARY OF DATA FOR STOCK PURCHASE FIVE DAYS AFTER RECEIPT OF VALUE LINE RECOMMENDATION Equally Weighted Portfolio Year 1971-1977 1972 Alpha% 4.6 7.5 t-statistic Beta 1.4 1.6 1.0 .99 1973 16. 2.3 .97 1974 1975 1976 12. -6.7 -10. 1.3 -.56 -1.3 .96 1.1 1.1 2.7 .61 1977 1.5 Value Weighted Portfolio Year 1971-1977 1972 1973 1974 1975 1976 1977 Alpha% 15. -4.9 15. 19. 14. 12. 24. t-statistic 3.8 -1.1 1.1 1.5 1.4 1.6 4.3 Beta 1.1 1.1 1.1 .89 1.3 1.5 1.4 Standard and Poor's "500" Year 1971-1977 1972 1973 1974 1975 1976 1977 Alpha% t-statistic Beta 19. -3. 14. 21. 20. 22. 4.3 -.68 .92 1.5 1.8 2.5 1.0 1.0 1.1 .86 1.2 1.4 33. 4.9 1.3 -68- Table 10 SUMMARY OF DATA FOR PURCHASE OF STOCK TEN DAYS AFTER RECEIPT OF VALUE LINE RECOMMENDATION Equally Weighted Portfolio Year Alpha% 1971-1977 3.3 1.0 1.0 1972 7.8 1.7 1.0 1973 16. 1974 1975 1976 1977 6.9 7.1 -12. -3.4 t-statistic 2.5 .75 .63 1.6 .71 Beta .97 .94 1.1 1.1 1.5 Value Weighted Portfolio Year 1971-1977 1972 1973 1974 1975 1976 1977 Alpha% 13. t-statistic 3.3 -6.1 15. 19. 12. 11. 21. -1.3 1.0 1.5 1.2 1.4 3.6 Beta 1.1 1.1 1.1 .89 1.3 1.4 1.4 Standard and Poor's "500" Year Alpha% 1971-1977 1972 1973 1974 1975 1976 1977 18. -5.0 14. 20. 18. 20. 30. t-statistic 3.9 -.97 .84 1.5 1.5 2.3 4.4 Beta 1.0 1.0 1.0 .86 1.2 1.3 1.3 -69Table 11 SUMMARY OF DATA FOR PURCHASE OF STOCK TWENTY DAYS AFTER RECEIPT OF VALUE LINE RECOMMENDATION Equally Weighted Portfolio Year Alpha% 1971-1977 1.6 1972 5.8 1973 1974 17. 2.3 1975 1976 1977 -11. -11. 5.2 t-statistic .49 1.3 2.6 .24 -1.1 -1.4 1.1 Beta .99 1.0 .96 .92 1.2 1.1 1.5 Value Weighted Portfolio Year 1971-1977 1972 1973 1974 1975 1976 1977 Alpha% t-statistic Beta 12. -12. 3.1 -2.8 1.1 1.1 20. 16. 10. 10. 18. 1.4 1.3 1.1 1.4 3.5 1.1 .92 1.3 1.4 1.4 Standard and Poor's "500" Year 1971-1977 1972 Alpha% t-statistic Beta 16. -12. 3.6 -2.5 1.0 1.1 1973 1974 1975 1976 21. 18. 16. 19. 1.3 1.4 1.8 2.2 1.1 .89 1.2 1.3 1977 27. 4.2 1.3 -70- confidence level, the probability of achieving it by chance alone is still quite small. This is the t-statistic for a -12.% alpha in 1976 regressed against the equally weighted portfolio. Otherwise the results seen are similar to the 5 day delay. Table 11 simulates a strategy of delays four weeks before action on Value Line advice. There are four negative alpha figures and two significant with t-statistics less than -2. The positive alphas are generally lower than the other delays, nevertheless they are primarily in the 10 and 20 percent range. It is shown that over the entire test period, results were significantly lower when there is delay in following the Value Line recommendations. However, even after four weeks delay, this table demonstrates that Value Line would prove useful in certain periods, although there were also significant negative alphas for the one month delay. The issue of the extent to which Value Line projects, and practices market timing remains unresolved. The methodology did not provide discriminate information for beta movement analysis. Although the tables demonstrate that betas vary yearly, there is no strong support for the intent of this variance. Table 12 supplies total period and yearly sub- divided one month Treasury Bill returns. Notice interest -71- Table 12 TREASURY BILL RETURNS 1971-1977 Total Period Return 40. Yearly Returns 1972 3.73% 1973 6.43% 1974 7.69% 1975 5.78% 1976 5.10% 1977 4.89% -72rates were highest in 1974 ranking at 7.7% and lowest in 1977 at 3.7%. Table 13 presents yearly cumulative growth of market returns, Treasury Bills and the Value Line portfolio. The choice of investment which would have produced the best results, at year end, independent of risk, may be determined in this table. From November 1971 to the end of 1972, an investor would have done best if he had invested in a value weighted market portfolio, although the difference between any of the market alternatives were within a few percentage points. To the end of 1973, the Value Line strategy would have been the best decision with growth of 1.18 times the initial investment. By the end of 1974, because of a dramatic market decline, one would have done best by having initially invested in the risk free asset, Treasury Bills. Of course Treasury Bill returns may have been higher if longer maturities were purchased. For 1975, 1976, and 1977, it appears that the initial investment in Value Line would have been the best investment, demonstrating growth to 2.2 times, 3.6 times, and 4.7 times. Table 14 presents market yearly growth. Volitality, can be observed, as well as the mixed returns of 1977, "zero" days, trading delay with the equally weighted index down 1.7%, Value Line up 29%, but the value weighted index -73- Cd ri a) HCm > a) H 0 N H H rl H rl H I- H * * * H H d(N (N U H r4 O N (U sr d4 4 -1 ¢) 1 >(a t-I I~ w zH ' UI CdH a)H zEm r-H Q)~ k4 -H Er' ;_I O H r-q O H r4 H rl O N C4H (N * * H H r CO m d * * H Hl zZ 4 3 E- N P; E-l p H a) Cd E- zH 0 ;4 3W O W zH > H U' ft: U- CD 0 _ Z CD O H E-4 I 03 1) >P IY FC2 1% ICO Ltl s-I U= t4 00 o d4 ro= % N dl dO -i O cn 12 - ) C .H ~v O P d 0 a I o r (dQ) O rN : oC C' ' CN H H H 35~~~~~~~~ H H H N OL C C) m N N N N (N N Ni NI H HP ) H H (N H O o C N r NY N - N no · rl o N o N ro ro II H r' o C C HH 0 o o o Q cn >4 LU rd fi IQ > -74Graph 3 CUMULATIVE PORTFOLIO GROWTH Treasury Bill = T.B. Value Line = V.L. Equally weighted = E.W. Value Weighted = V.W. Cumulative growth 6 4 3 2 r 1 1971 1972 1973 YEAR 1974 1975 1976 1977 -7 5- Table 14 YEARLY GROWTH PORTFOLIO (-5 Days) Year 1972 Equally Weighted 1973 1974 3.0% -34.% -27.% 1975 1976 66.% 56.% 1977 16.% ( 0 Days) Year 1972 Equally Weighted 5.3% Value Standard & Weighted Poor's "500" 14.% -20.% -31.% 31.% 26.% -3.6% 17.% -20.% 58.% 49.% -22.% 31. % 18.% 1977 17.% 1974 1975 1976 3.3% -26.% -16.% 62.% 43.% 1977 17.% 1973 (10 Days) Year 1972 1973 1974 1975 1976 1977 Equally Weighted 0.24% -20. % -12. % 66.% 39.% 17.% -11.% Poor's "500" 1974 1975 1976 1972 -33.% 27.% 18.% Standard & -17. % Equally Weighted 17.% -16.% -5.1% Weighted -25.% ( 5 Days) Year 12.% -20.% Value 1973 -1.7% Value Line 16.% -17.% .24% 27.% 11.% -8.5% Value Standard & Weighted Poor's "500" 90.% 71.% 34.% Value Line 17.% -8.1% 1.0% 79.% 54.% 29.% Value Line 16.% 15.% 11.% -21.% -22.% -12.% -24. % -27.% -5.% 35.% 31.% 7.8% 63.% 34.% -7.5% 20.% 15.% -0.78% Value Weighted Standard & Poor's "500" Value Line -15.% -16.% 8.0% 4.0% -23.% -26.% -5.6% 15.% 38.% 12.% 0.19% 15.% 34.% 5.0% 63.% -6.3% 19.% 25.% -76- Table 14 (cont'd.) YEARLY GROWTH PORTFOLIO (20 Days) Year 1972 1973 1974 1975 1976 1977 Equally Weighted -9.1% -15.% -3.% 59. % 29.% 18.% Value Standard & Weighted Poor's "500" Value Line 7.7% -14.% -16.% 32.% 8.0% 1.1% 8.3% -3.% -16.% -18.% 1.6% -0.08% 27.% 55.% 1.4% -5.1% 20.% 19.% -77- down 1.7% and the Standard and Poor's "500" down 8.5%. The empirical data presented in these tables provide strong support for the verification of Value Line Investment Survey as a useful investment advisory service. -78Chapter 6 TAXATION AND TRADING STRATEGY Taxation is an important determinant in the investor's behavior. This section is intended to be thought-provoking, however it is not a comprehensive dissertation on tax management in investment strategy. People often talk about taxes, they usually complain about paying taxes which are too high. Frequently, they look for a way to minimize their tax liability; however, they continue to make investment decisions with little regard to the future tax consequences. In my classes at Sloan there are many instances where I have been part of purely theoretical decisions where taxes and transaction costs are ignored. I have also wit- nessed taxation discussions which ignore the advanced management techniques necessary to do sophisticated financial planning. As C. P. Snow would say, I hope to bridge the gap between these two extremes and provide an investment frame- work which considers with a careful methodology the consequences due to taxation. Textbooks seem to skim over the issue of personal taxation, yet they describe elaborate methods to maximize financial decisions. every one There are also books describing almost of the myriad of taxation issues -- yet no one -79speaks of maximizing decisions after taxes. I admit my methods may show rough edges, but my goal is to examine tax strategies to maximize return after taxes. I assume that an individual's primary goal is to maximize his economic welfare when involved in an investment strategy. taxes. This maximization to be realized must be net of My purpose is to develop a simple model example of taxation which takes explicit account of the individual's liabilities both for ordinary income and capital gains taxes. I hope to identify a set of guides and logical investment decisions. This section will compare two stock trading strategies with the goal of maximizing profit while minimizing taxes. This is a study of a complex issue with a myriad of possible solutions. If I omit analysis of a situation it is usually because the analysis would be self cancelling -- I plan to attack those areas which will express differential results. My method will employ a scenario of "what ifs" and apply this guideline to the strategies I study. I plan to discuss the following propositions: -- What is the Value Line trading strategy, and what are the tax consequences of the strategy to the individual -- investor? How does a similar analysis of the Standard and Poor's "500" stock index hold up? -80- How -- are the returns distributed between dividends and capital gains? How do taxes impact upon market gain which is real gain, and extra gain due to the ability to pick stock by active trading? Does the government, by taxing gains and subsidizing losses, reduce risk to the individual investor? How to manage taxes to minimize them. Look at preliminary data. --Ignore the approximately <1% stock transfer taxes. When Value Line publishes a return for securities in a portfolio, they ignore brokerage costs, dividends, stock transfer taxes and all consequences of income taxes. The Group 1 portfolio I will be discussing for the remainder of this chapter will be an equally value weighted portfolio of only New York Stock Exchange Rank 1 securities. Weekly the portfolio will be updated and theoretically rebalanced to maintain equal weighting. What tax issues relate to these tax trading strategies? The major point of interest is the differential treatment of long-term capital gains and ordinary income. Congress has extended special advantageous tax treatment to capital gains. The House Ways and Means Committee has -81- recently passed an amendment to reduce the capital gain tax to provide greater incentives to invest in growing companies. The May 8, 1978, Wall Street Journal included "the reduction in the maximum rate on capital gains taxes from the current level of 49.125% to 25% in 1980 would have a positive effect on economic growth while reducing the federal budget deficit." What was the law in 1977 regarding the taxation of capital gains? It was that one half of a long-term capital gain is included in an individual's income tax base, providing for an effective tax rate from 7% to 35%, according to a government publication, the President's 1978 Tax Program. There is a special rule which provides that 50,000 of these gains each year are not to be taxed at over the 25% rate. 2 How do the capital gains rules affect our portfolio decisions? Our first realization is that we receive income from both capital gains and dividends. Dividends are received while the stock is held in the portfolio and are taxed at ordinary income. Capital gains are taxed at a lower rate if the stock was held long enough to qualify as a long-term gain. The individual's portfolio should be managed to take advantage of long-term capital gains. This is a divergence :from just following Value Line recommendations. In 1977, the gain or loss from the sale of stock held for more than nine months was long-term. In earlier years the cutoff was -82- six months, and this year and in the future it is 12 months. Net short-term gains, or the excess of net shortterm gains; over net long-term loss is taxable as ordinary income, according to a November 4, 1977, note in Value Line. Taxation of long-term gains is explained above, however, the untaxed half of the gain is treated as preference income and may be subject to minimum tax. If tax preference for the year exceeds one half of the individual's regular income or $10,000, whichever is greater, the excess is taxed at 15%. Net short-term capital loss, or the excess of loss over net long-term capital gain are deductible from ordinary income starting in 1978 of $3,000 per year with the excess carried forward as a short-term loss carrover to future years. Further, one half any net long-term capital loss is deductible from ordinary income. The portfolio may not commit a wash sale, that is, a loss not allowed for tax purposes, if the same security is sold and bought back within 30 days. This dictates that if you want to take a loss you must double up on the security you want to remain in the portfolio. A possible strategy for an investor to gain tax advantages while trading under the Value Line strategy would be for the individual to realize losses to offset gains before the end of the tax year. Then these stocks could be brought back into the portfolio if they were still -83rated Group 1. This sounds simple, yet taxes are being deferred. When the government taxes gains and allows loss deduction it is essentially reducing the risk of any investment you make. Because of this risk reduction, individuals may be willing to hold more volatile portfolios than would otherwise be the case. ]n "The Implications of the Capital Gains Tax for Investment Decisions," 3 by Holt and Shelton, the being locked into an investment by the capital gains tax is discussed. The article was written in 1961, and recently the law has been changed; the capital gains tax can no longer be escaped by dying before selling. The new law values trading gains against the stock price as of December 1, 1976. They ask how much extra yield is necessary in an alternative investment to overcome the disadvantage of incurring the capital gains tax. In this study capital gains tax cannot be avoided, only deferred. As long as it can be deferred it representes a free loan from the government. When following Value Line's recommendation to sell a stock and replace it with a new one the following question should be asked. Are the future dividends and capital gains on this new choice sufficient to overcome the amount lost in capital gains taxes? The article concludes that this yield differential is smaller than most investors realize -- often -84on the order of 1%. So capital gains taxation may not be a major implication while following the Value Line system, although as we will shortly see, much of the Value Line portfolio does not qualify for long-term capital gains treatment, due to the high portfolio turnover ratio. My next step is to examine the actual Value Line portfolios. The specifics are chosen from available data, which is really not in the proper form for this analysis, so many assumptions will have to be tolerated. I studied the most recent 21 weeks of data -- conclusions were drawn from the period of August 5, 1977 to December 30, 1977. At times comparisons may seem out of time frame -- and still, within a broad tolerable range this analysis is still useful in an educational and thought-provoking process. Black's study "Yes Virginia, There is Hope; Tests the Value Line Ranking System,"4 ignored tax consequences. Black revised portfolios on a monthly basis, while I am employing weekly updating, which might account for the higher turnover observed, or we may now be in a more volatile market (Figure 15). In Black's study extra return was measured at 10% per year with a "T" value of 4.0 a significant result. Diversification with the market showed a correlation coefficient of about .95. Excluding transaction costs and taxes, Black's results may seem unrealistic for the investor, especially -85Table 15 APPROXIMATE COMPOSITION OF VALUE LINE PORTFOLIO Rank 1 22% of stocks qualify for long-term capital gains 78% short-term capital gains Average price of stock Average dividend Turnover Rate 25.54 1.11 or 4.35% According to Black study 130%-88% According to new data 225%-140% Black found about 10% excess returns on Rank 1 7% excess returns on Rank 1-3 Hold Our data is from December 31, 1976; December 30, 1977 Value Line August 5, 1977; and -86- because of the 130% turnover ratio in the Rank 1 Group. This high turnover leads one to believe that the majority of the individual securities are held for less than one year, the length of time necessary to qualify for long-term capital gains. In my new study, it was found that turnover was closer This is without to 225% creating large transaction costs. the extra trading which would be necessary to maintain an exactly equally weighted portfolio which is an assumption which I will now relax. Turnover also increases because of the large number of rank changes when quarterly reports are issued. Methods may be employed to reduce transaction frequency. These include selling a stock only after it falls to a Rank III, as suggested by Black. This would also tend to increase the probability that a stock will qualify for longterm capital gain treatment. volatility Using this strategy, Black's of turnover was reduced to 88%. According to my translation analysis I would predict a 140% turnover. Under this newly defined strategy extra return was also reduced to 7% with less trading, as reported by Black. Now let us take a broad look at the application of the rules of taxation to the Value Line strategy between December 31, 1976 and December 30, 1977. 22% of the stocks ranked 1 remained in the portfolio, and therefore qualified for long-term capital gains treatment. Therefore, approxi:mately 175 stocks were traded in and out of the portfolio -87- during the year -- all representing short-term gains. As a statistic portfolio for snalysis I chose August 5 1977. 78 of the 100 stocks in this portfolio are traded on the New York Stock Exchange. this portfolio was $25.54. The average price of a stock in The average dividend was $1.11 or about 4.35% of the price of the stock represented yield on an annual basis. According to Value Line the average percentage change in price between December 29, 1976 and June 29, 1977, for Group 1 stocks with weekly updates was a positive 15.4%. Approximately 13% of this gain is from stock appreciation and about 2.4% from dividend income. I also attribute about 22% of the capital gain to long-term capital gains under the one year holding period rule by extrapolating made above. the assumptions On an annual basis the gain distributes as follows: Assumption #1 Long-Term Capital Gain 6.% Value Line Rank 1 Short-Term Capital Gain 20.4% Weekly Trading Dividend-Ordinary Income Total Annual Return 4.4% 30.8% This result is due largely to a very good time period for the market. income. Most of the gain is taxed as ordinary Would it be advantageous to try to hold stocks longer to take advantage of the favorable long-term capital -88- gains ratle? This would reduce transaction costs conceivably. What would the situation look like if we viewed the same portfolio and time frame as under Assumption I, with one change, the restriction that only one portfolio change is allowed halfway through the time period at the six-month point. In this case the six-month gain was reduced from 15.4% to 10.1% according to Value Line. My estimate of the turnover ratio under this assumption is about 50%. This is only a ball park figure and verification requires data collection of stock tracking them from Rank 1 all the way until they enter Rank III. This is beyond the scope of this thesis, so my 50% figure must suffice as an estimate. conditions Assumption Under these II is created: Long-Term Capital Gain 7.9% One Trade Only Short-Term Capital Gain 7.9% Rank 1-3 Value Dividend Income 4.4% Assumption II Line Total Annual Return 20.2% Further insights may be provided by actually taking a dollar position in each of these portfolios. A hundred share of this portfolio as is on August 5, 1978, would cost $2,554 in December 1977. (There are brokerage costs of about 3% and stock transfer taxes which I have not included. Also, the portfolios may not be valued exactly -- but the approximation is sufficient.) under Assumptions I and II? How do tax consequences vary -89- Before attempting an illustration I must restate the simplicity of this analysis and point out it is only for illustrative purposes of possible tax consequences. It is unusual to examine a period of such high return, which do provide us with capital gains and not losses, after all I have not studied securities individually. Another simplification of this analysis which may seriously bias results is the lack of a measure of capital losses. This is an important fault because the losses could be directly subtracted from gains for tax purposes. Assumptions must be made about our hypothetical portfolio holder. They are as follows: 50% income tax bracket, with less than $50,000 in capital gains so that capital gains are taxed at 25%. portfolio Another assumption of the size of $11,500 provides for dividend payments of approximately $500. which, taxed at 50% as ordinary income before the $100. dividend exclusion, creates an effective tax rate of 40% on dividends. The net gain after taxes on Assumption I was $1,989.50 or a rate of return of 17% compared to a $1,435.00 gain under Assumption II representing a 12% after tax return. So, at first glance it appears that it was not worthwhile to reduce portfolio turnover to seek capital gains preference tax rates. However, this is not necessarily the case. The increased trading costs associated with Assumption I could -90easily consume the 5% differential and make the two assumptions finish in a dead heat. There is still another comparison I would like to make before concluding this glimpse into taxation study. That is a comparison of Rank 1 Value Line stocks from 19701976 to the Standard and Poor's "500" composite index for the same time period. The Standard and Poor's "500" stock index contains no trades, and therefore no trading costs for the entire period. Essentially, I am comparing a buy and hold strategy to active portfolio management. Again, unfortunately there is a major drawback to this analysis, and that is that I do not use a valid risk adjustment factor for Value Line, which may tend to overstate its returns. Results for Standard and Poor's "500" show only a 27% after tax return from 1970-1977 hypothetical dividends (July). Using the same $11,500. investment and applying a 40% tax on and a 25% capital gains tax, (actually I have omitted to account for the time value of dividend payments made seven years ago), however, since I am ignoring this dividend reinvestment for both Standard and Poor's "500" and Value Line, I believe the bias will remain small. The total return for the entire period is 27% after tax for the Standard and Poor's "500" and 114% for Value Line's data. However, this comparison does not take account of the -91transaction costs of the 13-225% turnover of the Value Line portfolio. To simulate this, I will recompute the Value Line returns reducing each yearly yield by 5% to account for the transfer tax and brokerage costs of such heavy trading. After the recalculation of Value Line results, reducing the yearly returns by 5% yearly, the pretax profit is reduced to $7,880. which, after taxes, reduces to the following: Long-Term Capital Gains (22%) 1733.60x25%= 433.40 Short-Term Gain 6146.40x50%=3073.20 Dividends 5136.11x40%=2054.44 Total Tax 5561.04 Total Gain $7455.07 After Tax Gain 65% I believe this to be a more realistic figure for the real Value Line return, which is still about two and a half times greater than the Standard and Poor's "500". In conclusion, I have presented two views of after tax returns following the Value Line system. In each case the gain in return due to capital gains treatment was balanced against a lower excess return, or index, involved with the one year holding period. In all cases, the probability of error in analysis in this taxation section is high, due to high portfolio turnover, transaction costs, -92offsetting gains and losses, and the capital gains holding period. When a more suitable data base is complete for taxation analysis, a complete story can be devised with more precision, less restrictions, and fewer assumptions. This is an area which is ripe for further study. In the last few weeks the 1978 tax law revisions have already altered the analysis applied to the 1977 strategy presented here. Essentially stock investors have gained. The new Law liberalizes the rules regarding capital gain. "The maximum rate falls from 49% to 28% once you pay taxes on 40% instead of 50% of long-term gains. ...15% minimum tax no longer applies to the untaxed part of capital gain." This is a major challenge to investors since the rules have been so dramatically changed. In a nutshell -- there is less of a deterrent to selling stocks which show capital gains. is now However, the holding period for capital gains 12 months. 6 -93Chapter 7 CONCLUSION, SUMMARY AND TOPICS FOR FUTURE RESEARCH Evaluation of investment performance is often considered an after-the-fact measure. However, it should be part of an on-going improvement process in state of the Investors who pay the costs associated art finance theory. with active portfolio management deserve to have a true performance measure. At times it may be difficult to separate performance due to skill from that due to luck. The importance of these findings rest on Value Line's expost alpha values, i.e., the vertical intercept obtained when an expost characteristic line is fitted using excess returns in an ordinary least square regression. Value Line's expost alpha should be in- terpreted as the average difference between its return and that of a passive buy and hold market strategy of equal risk. This thesis has examined Value Line's historic alpha values. What does this imply about the future? The significantly high t-statistics demonstrated by Value Line data over many consecutive years demonstrates consistently superior performance which can be projected into the future. The objective of an actively managed portfolio of common stocks is to choose securities so that a greater return results than that produced by an index fund or a naive investment strategy. The actively managed portfolio must -94- overcome additional costs not incurred in a passive portfolio strategy. These include the cost of gathering and analyzing information, transaction costs of executing trades, nonoptimal taxation decisions, and non-efficient portfolio diversification. This study analyzes the 100 stocks Value Line ranks "1" for year ahead appreciation. Transaction delays are computed for -5, 0, 5, 10 and 20 days. criteria of security evaluation are: Value Line's four main non-parametric value position, magnitude of over or underevaluation, earnings momentum and an earnings surprise factor. In previous Value Line research, Shelton1 found investors displayed superior performance than efficient market theory would suggest. Hausman2 points out flaws in Shelton's statistical tests and recommends additional study. Murphy 3 essentially agrees with Shelton, discovering significant results consistant with the random walk. Conversely Kaplan and Weil's 4 results support the efficient market hypothesis. They believe that the Value Line rankings are flawed and that most performance variation is due to stock risk differentials. Black's5 study concluded that even with the imposition of transaction costs,-Value Line continues to give significant positive results over a five year period. Fama6 provides strong support for efficient markets in his study of the 30 Dow Jones industrial stocks from -951957 to 1967. Sharpe,7 studying 34 mutual funds and Jensen, 8 studying the performance of 115 mutual funds each lend support to the view that capital markets are efficient. This Value Line data was hand collected covering the period from November 5, 1971 to December 30, 1977. Computer programs were written to: folios; 2) 1) generate weekly port- name and verify portfolios; 3) compute individual portfolio and market returns for certain trading day delays; 4) calculate Treasury Bill returns; and, 5) calculate ordinary least square regressions. Market indices studied included an equally weighted and value weighted New York and American Stock Exchange Index, as well as the Standard and Poor's "500" composite. In the regression analysis the Value Line excess return was the dependent variable and the market index excess return was the independent variable. Ignoring transaction costs, the Value Line portfolio would have grown to 5.8, 4.8 and 3.1 times the initial investment for trading delays of -5, 0, and 5 days respectively. Each of these is superior to the best of the market indices, the equally weighted index which grew 2.3 times. A similar investment in Treasury Bills would have grown 1.4 times. R2 terms were all high. Executing market action on the Friday of publication the Value Line portfolio when regressed on the equally -96weighted portfolio resulted in a positive alpha of about 12% with a t-statistic of 3.6. This result strongly refutes efficient market arguments and suggests that Value Line recommendations do produce investment strategies which consistently outperform the market. This figure is large enough to absorb considerable transaction expense. Assuming an individual's investment goal is to maximize final economic wealth, consideration must be made for taxes and transaction costs. This requires balancing extra return against capital gains holding periods, and reduction of the agressive Value Line trading turnover rate. The rec',._" major tax revisions provide new opportunities for further investigation. Many unanswered questions remain. Is Value Line analysis biased towards low price/earning multiple stocks, or do they favor smaller, inefficiently priced companies? To what extent do they rely on security analysis, or on adjusted beta and market timing? Further work would also be valuable in studying the optimal investment tradeoff between superior returns and transaction frequency. -97Appendix I CUMULATIVE COMPANY LIST -98FILE: C-L"'UL 1 = 2 = . 1 L I = 235 = 2 8 5 = 1 6 = 7L5 :10 7 = 745-51C 9= 10 = 11 = 31C J1 - S 814 .1 867710 92661C TI D 4I NSCN H ALAN ALA ALRCRTSONS 17 = 131 3410C 13788F31 13 = 1 3961 19 = 21 = 147521C 1717, 1 173721C 122 = 1 763413 15 = 16 = 20 = _23 = F CORP AS ALCC STD LL Cumulative company list. TIIC CORP LaPS I JC ALFXAND E S 710 la2 '51 CO 5 OOD STL O0 A. I NTST CO L-ALCN 1908 CO AT IRORNE F HT CORP AIRCO INC 116t01C 117341C 14 = -3HA4 Y INC CRP -ALLrGHENY LUDLUM ALL-N4 SR CUD INC ALL IED CHE M CORP IND 24 = 25 = ALL IED LS 194111S ALL !ED PROCS 26 1 53 7 10 ALLIED SUPERMARKETS 27 = 1 3451C A L LIS 23 227711C = = '29 = 30 31 = 231411 = 235 1 13 235511C 33 = 237531C 34 = 2377110 = 36 = 2473 51C -3 247531C 39 = 40 = 41 = 42 = 43 = 4i = 45 = C 2553710 2 58161 250731 4.7 = 266811C 4.P = 268I7 q IC 49 = -' 733 2C 50) = 27421 9 274471C 2762 71 53 = 54 = 55 = ' I C AN T AM ER ICAN BAKERIES 2926710 C 2946510 2960 1 ' C BRO.DCASTIN A INTE CHAIN CAB CfONSUMER IN CYANAMID CO 2532110 46 = = = A M _ I CAN AME' ICA-I A 2535 71C 265731 0 51 52 IN C A1CCR D I NC BLC G 251051C 252311 DE AMERADA HESS CORP A'E.9ICAN AIR FILT ER IMERICA^N AIRL 239D41C 2 40'6 q 1 37 = = CORP AMBAC INES INC AMEDACE CrRP 32 = 35 INC CHALME RS O RP ALPHA -PORTLAND IN DS AMALGAMATEC SUGAR C 227711u ONiITOR SYT q? EN RAL TCP AHM 9z2921C L TDV'"NCE INVS CORP AE OJET EX ' C P AETN'A LIFE . CAS CO ALAAMA = T I' ,% T n Iq 1028410 13 S I IC I>NOS ADAVS EXPRESS 1 77,8;' 12 = J AD 4 CNV P A 71 2)241 3 9 = TA A1ME ICAN AM ER ICAN AMER ICAN AMEE A;M CTA4N ICAN AM E I CA N ELEC PWR EXPRESS IN CO FAMILY CORP GEN INS CO HCI ST DE R HCSP SUPPLY IlNVT CO M'AIZE PRODS 41 , , ICAN C AN AMER A MEI A.r: II C AN CAN E:- INTL IJ ECIORP IN MTRS COR P !ME? ICAN RESH r EVC At~E I CAA AMER ICA I SE.ATING CO AME2 .I ANq StHIP BLDG C 56 = 3 071 01 3114110 60 = 61 = 62 = 3 = 64 65 C;' C C- C. K1 K'.j C. = 66 = 0' INC 321771C 3391 3233910 = 336091C = ANDERSON 3531301 = ANKEN INCS 321721- 1C = MTL = APACHE 'ORP I NC C CLAYTON CORP 37411 68 69 = = 3751910 3752 1C = APCC OIL CRP = PECO CORP 70 = 393751C = ARCATA 71 72 73 74 75 = = = = = 3:94831C 4'055510 4208312 4219510 4232110 DANIELS MIDLA = ARCR CO = ARIZ7CN oPU SV = RrMA)A CP = ARMCO STL CRP CO CORK CRUMSTRONG = 76 77 78 = = = 4246510 4333 91 445401C = Ar"-STRON0 RUBR CO = ARV N I iS ¢ INC = ASHLAND CIL INC ATL CORP 79 = 80 = 81 = = ATHLONE INDS INC 4748310 = ATLAS CORP 492673 10 = -ÛSTRAL CIL INCJ 5251 82 = 330151C = AUTO"ATIC DATA PROCE 83 = 34 = 5321310 = AUTCVATICN 535011 r = AVCC CORP IDS 5 = 536271C = AVERY ITL CORD 86 = 538071C = AVNET 87 = 5499710 = AZTrC OIL 88 89 = 5f147 5171C 90 = 91 = 92 = 5635710 5725510 6C2211C 671491C 393 = 98'R6911 = INC INC GAS CO WILCOX CO ACCCK = BACFE GRCUP INC INC = bAKER IDS = BJANGOR PUNTA CORP = ARRBER OIL CORD = BASIC I1 MFG INC 705911 = BATES 71.7071C 71.892173;23910 LCVB INC = BAUSCH = BAXTER TRAVENOL LABS = BAYUK CIGARS INC 98 = 7588710 = BECTON DICKINSON 99 = 766351 = BEECH 96 7 = = = 77i'41910 = 3ELCO IRCRAFT PETE CORP CORP 7745510 =BELDEN 77'4110 = ?ELDING 103 = 778511 = PELL 104 = 814371C = GEMIS INC RKEY PHCOT INC = 24491C INC = °EST PR OS = ,ETH-LEHE!' STL CORP 101) = 102 O~ FCC = 1C0 C,, = 71r 67 94 = 95 = C = = AMPCO PITTSnUr" = AMSTAR CORP I7NDS INC = AMSTED 323 ATI 9CCOIVERS A= 7,=EICA'"I1ST" IrC I STERILIZE . AMEICA C AN S TOE = A E9IC = A1EcC, I Co 2q7171710 3 -8710 301 -951C 59 1( = 57 = 58 = -99- A LISTm!R CUL"liUL FILE: 105 = = 106 = 107 = 10- : 109 = 110 = ~44191) 5511 8f6, 875091C 92510 = RL~I 935451 C = ,LI S = BL3CK 936711C CORP HEMINWAY INC HCWELL CO JOHN C ' LAUGHLIN R INC H I C L CNIT3R SY' -100FTLE: CULMUL LISTPR A .:' V=RSATIONY^L C 111 112 C 113 114 115 116 117 118 = 5293l* OLU- -= = 6 77 9 1 7C0231 ?On = 973831C = 9 5451 = 1020 9710 = 1054251C = 190C431 1 1Q = 120 121 = 122 = 123 = 1 10' 971 C O0IOE 3K BOURNS INC ?RAiIFF INTL CORP STPATTON CO _nICS ORI STOL 'YERS CO BROWN CO 3RO,N FORMAN DISTILL RRUNSWICK CORP BRUSH WELLMAN INC 124 = 11883510 = 11?00710 C 15291C = 1206553C = 12237510 = 12316 91 13C = 12l BUD CO 3UDCf-ET INDS BUNK<ERRAMC CORP 3URNS C t ., C C C C C© C, C, 153 - 15 1 = 15C431C 152 153 154 155 156 157 = 1513031C = 1523121C = 1 536091C 158 159 I233 1= 530:93 3 11C = 153 971 C = 1551771C = 1 554471 =- 56,4n210 = 156 82510 = 1568791C 161 = 1 5 50 : 1 0 162 = 15852513 163 = 16278910 164 165 = 16326 71 = 163603 1- I CORP L C AMER C 133 = 12705510 134 = 1276951 135 - 12P7931C 13021 7 1 137 - 13106910 139 - 13441110 139 = 1344291C 140 = 13 644 93 141 = 139'86110 142 = 14233910 143 = 14414 1 1 C 144 = 14423851 145 = 14446510 146 = L445011C 147 = 1462851C 148 = 14E42910 149 = 1491231C INTL SEC SVCS 'USH UNVL INC 841, =125011C INC OUFFALO FC7GE CO 131 = 1256151C C/ CASCACE CO'P UNTtH CLUJ INC 11563720 1170431C 11742110 125 126 127 129 129 C INC 11 5 3 3 11 C 132 ELL TE BRCOKS INC T INC S CORP CAB T CORP CAESARS WCRLD CALDOR INC INC CAL IFCRIN IA FINL CORP CALLAHAN MNG CORP CA"PBELL RED LAKE MI CrAMPBELL SOU ° CO CAN4ADIAN PAC LTD CAPITAL CITIES CARLISLE CAROLINA COMMU CORP PWR & LT CO CARPENTER TECHNOLOGY CARRIER CCRP CARIERS CARTER £ GEN CORP WALLACE 1",C F COOKE INC CATERPILLAR TRACTOR CECO CORP CELANESE CCRP CENCO INC CASTLE C EN TEX CENTRAL CENTRAL CORP HUDSON GAS & LA ELEC INC CENTRAL SOYA INC CE N TRAL TEL £ UTILS CENTRNICS CATA CCMP CERRO CORP CE-RTAI TECC CORP CHA'OION HO'E BLORS CH 'P IOl INTL CORP CHECKER VTRS CORP CHELSEA INCS I*IC CHE v"TRON "CRP 'CNITOR S, -101FILE: C C 166 167 = 1651591 = 1 7 2 6 31 CHOC K FULL 166 169 = ) 1 17113o .HC;.rnALL3Y = 17111IhC CHRYCLER 170 = 1715831= CH"CHS 171 = 172 17 = 17? !73 = 1778461 = 1 814861 174 = 175 C C C C 1 94861C = 18937313 19055610 176 = 177 178 = 19325210 = 19337310 179 180 = = 181 = 143551C 19484610 188 = C C "' C C C C = CLARK = CLUETT = = C'-OLONIAL PENN GROUP IN CO.MUNICATI 0Ok'BI1ED = COMRUSTICN = = 22065110 = 2-D490 D13 23536313 20681313 20719212 COMODORE = COMPUTER = CONE MLS CCRP SCIENCES nORP CO CONGOLEU. CRP CONRAC CORP CON SOL IDATEC FREIGHT = CONTINE"TAL AIR LINE 107 198 21991 200 291 = 21079510 = 21129 1 = = 2113271 = 211 4851r = 21168710 232 = 2120751C 203 = 21229312 EQUIP ASS SOLVENTS CO'MERCIAL OIL REF = COMMONWEALTH = CO~~~rUNICATICNS SATEL = COMPUGRAPHIC CORP = = COPPER CORP CONTINENTAL ILL CORP CO>,TINENTAL .'MTG INVS STL CORP CONTINENTAL TEL CORP CO'TINENTAL CONTROL CATA CORP DE COOPER I NDS INC COOPER JARRETT INC CONTINENTAL = CONTINENTAL = = 21236310 = = = 21666210: = 2166871 = = 2167051C = COOPER LABS INC = 21683 10 = COOPER TIRE S RUBR C 209 = 21721 10 21C = 2175251C 211 = 21768710 212 = 21866 11 213 214 215 216 = 217 = 21032710 = 2202911, = 22374110 = 2242031n 224391 = 22781310 2194 = 2286691 23003251C 220 218 r CO = COLT INDS INC DEL = COLU?'BIA PCTURES = 2)92911C = 2092371C 206 297 208 PEARCDY = SCOACHMENINS INC = COASTAL STS GAS CORP CO = COLDWELL BANKER = COLE rO INDS INC = COLE2AN INC I FOOCCCS INTL = COLLINS 105 196 205 REFNG SCO OIL 2023811: 204 C VILACROC TIVESTING CC = : 2034171C ,· CORP RIED CHICKE I'4C INN\ATI CITY CC AERN COR 2002911C 193 I 22 = "0R P VA C NTS = = 184 = 20320113 19 191 = CHESAP-K 19501 810 = COLLINS RADIO CO 1958461C 13 = 19686410 134 = 1982791C 185 20010113 182 186 187 C. 0C"V A L7TST[R C'L'LUL = COPELANDCRP = COPPER RANE CO = COPPERWELD CORP = CORCURA CRP = CORNING 3GL.SS WKS G BLACK CORP = COROON ICATION = COWLES C.UN COX BROACCASTIG CCR CO - CRANE = CROUSE HINDS = = CROWN ZELLRBACH COR SULLIGAN ITL CO O SAT !7 "4'A ,, '~ U'T, T AR ' -102FILE: f ?- C L",.UL LI STDR 221 = 231021 1C 222 : 23112931 C ( C r C .* C C -F; "i "-:··-I ,· i. -··'4 71.-j i i CC. C C. C6' 4E I jr t[ I CRU(G STSR 225 = 2322191 CUTTER LS INC CYCLOPS CCRP 227 = 2332911 22q = 2333511 229 230 231 232 233 234 235 236 237 238 239 249 = = = = D P D W F G INC CORP 2357731C DAN RIVER INC 23581110 23742410 2376881C DANA CRP DART INDS = 23810020 INC CRP DATA4 GEN 2 3 8 1071C DATAPOINT CCRP DATAPRODUCTS CORD DAYTON HUDSCN CORP 244 1 9 91 DEERE = 23975310 CO = 2452 171. DEL 24783 2C = 2486311C DELTEC ITL LTD ENG DENNISON MFG CO = 2487931 N'ONTE CCRP ;ENI'IJYS INC 2503611C 25059510 2521651C DESERET PHARMACEUTIC 25243510 DI 25246810 25274110 25357910 2538491C 249 = 25411 11C 250 = 2546871C DIAL CORP DIA M OND SHAMROCK COR DICTAPHONE CORP DIGITAL EOUIP CORP DILLINGHAM CORP DISNEY WAL T PRODTNS 251 = 25612 1C DR PEPPER = = 24 3 = 244 = 245 = 246 = 247 = 248 = 242 C, '! , ,F I S CU'INTINliGHA UPrTISS WRIGHT CORP ER INC CUTLER H 24 1 C ' C:,V 223 = 3156 11 224 = 2321651C = 2325251C C A 2_52 = 2570751C 253 = 25709310 254 = 25714710 255 = 2578671C 255 = 2582371C 257 = 2583631C 258 = 26000313 259 = 262181C 260 = 26774110 261 = 2684571C 262 = 26915 730 263 = 27033010 264 = 2761911C 265 = 27646110 266 = 27874910 267 = 2787851 268 = 2 30751r 269 = 281 691c 27C = 28336210 271 = 2R55511C 272 = 23643420 273 = 2911C 011C 274 = 29117310 275 = 29121 1C DE SCTO INC DEXTER CORP IORGIO CORP CC DOME MINES LTD DOME PETE LTD DOMINICK F IC DONNELLEY R , SONS DOR IC CORP DORR OLIVER INC DOVER CORP DRUG FAIR INC DYMO IDS INC E o C G INC E SYS INC EASCO CORP EASTERN AIR LINES EASTERN GAS IN FUEL A ECHLIN MFG CO ECKERD DRUGS INC DEL EDISON BROS STORES I EDLaRDS A G C- SONS P SC CC ELECTRONIC ASSOC INC 7L LS IN ;IATL I 'JDS I] C EMERY AI R FGHT CORP EMEPy. 1>405 EMHAP.T CRP IS VA S T I N', L ' T -103ULM1UL FILE: 277 = 29171310 = 2928481C 276 C LI ST2?R 27'9 = 279 29% = 293151 = 294091"10 281 2Q2 = = 283 = 2933891,3 2944971C 29647t012 2966591C CCNVJRSATIN A = E4P I RE G.S CORP = -NSEHAR. II'INECALS T'41 BUSINESS FOR ' = = NTEY INC = E'Y IROTECH C'3 = ECU ITABLE GAS 0O = = ES"A.RK ESOLIRE INC ItNC 234 = 29742510 = ESTERLIN = 29765912 = ETHYL 286 287 298 289 . 20C = = = = = = = = = = FAIRCHILD CAMERA I FAIRCHILD INDS INC FALCON SEABCARD INC FANSTEEL INC CRP FAR WEST FINL 291 292 = 3073871C = 31322513 = = FARIANH FG I'NC FEDERAL CO 293 = 31354 294 = = FEDERAL 295 296 297 C 298 = 3136931 FEDEPAL PAPER BRD IN = F = = 299 = 300 = 31571110 3036931C 30371110 30608410 30726110 3073511' 91 31355610 31376510 = 3138551C = 31412312 3C01 = 302 = 3154051 3165491 1 3172971 303 = 31731513 3C4 = 3174951C t'C (C. C o: ~319 CRP 285 CORP = FEDE 1 AL FEDSAL CORP OGUL NATL RES TG ASS CORP = FEDERAL SIGNAL CRP = FEDERATEC EV CO = FERRO CORP = FIRREBOAR = = CORP LS FIELDCREST FILYWAYS INC- INC = FILTROL CORP = FINANCIAL FEDN INC 305 = 3183151C = FIRESTONE TIRE RUB 336 = 31944110 = IRST CHARTER FINL C 307 = 31945510 = FIRST CHICAGO CORP 308 = 32048810 = FIRST HARTFCRDCORP 309 310 FIRST MTG INVS PCRTER CO = FISCHER = 32104510C = 33759310 FCODS 3L11 = 33781910 = FISFER 312 = 33802710 C = 3390991C 313 = FITHR SCIENTIFIC CO = FLEE-TWOOCENTERPRISE INC 314 315 = 33937612 = 34063910 = FLEXI VAN CCRP = FLORIDA EAST COAST R 316 317 = 3410al1C = 34109910 = 318 = 34317210 = 34386110 320 = 34487210 321 = 34551410 = FLORIDA FLORIDA PWR LT PWR CORP CO = FLORIDA STL CORP FLUCR CORP = FOOTE CONE BELDING MC KESSON I = FOREOST = 322 = 34746010 = FORT HOWARD PAPER CO 323 = 3502441 324 = 35160413 325 = 3567151C = FOXPCRO CORP CO = FREPORT MINERALS 326 327 = 35937012 810 = 3 12 = = R1JEHAUF CCRP FUCJ.LA I11CS IJC 328 = 36142:1 *= G . F C~P 329 = 3616061 330 = 36464 '1 f = FOSTER WHEELER = F USIrNESS QUIP = SAM PLE SKCOSO INC C I 4L CP I TR < - 1FT ., `7 C tjL " U L '' "T "R ' C'L V "SATrA 0'L v . T. 1 331 = 3676221C = SATWVAY I NS 332 = 368f30212 = GE.7- A-'EvA 333 = 3691401C- GE ErAL BATTERY CORP = GENHAL 4ABL-E'SRP L INEMA CRP = ESE = -SE'RAL CYN.AMICSCOR 1 AL INC E L I NVS I 334 = 335 = 3693521 3 36 = 337 = 3700641 =- "ER 33 = 37701171 C 339 340 341 342 343 = = = = = = GENt RAL INS TR C P GENE .AL tiEO CORP = GENERAL MLS INC rF" 344 = 3713521C = 345 346 = 3724511S = 37'3291C = SENSTAR LTD = GEORGIA P40 COP = aGE%,ER PROS CO = GIANT PORTLA!4D M4S = GIBRALTAR FINL CCRP = ,IDDIGS LEWIS INC = GILLETTE CC = I OS C IC = GLEASON WKS ! C. C C; .. 35929 3 6955 1 370t2981 37033412 37'05141C 37062210 3708561C 347 = 37371210 348 349 = 37'453212 = 37'46581C 350 351 = 37504613, = 3757661C 35 2 = 353 = 37735210 37'610 1 C.. = GENERAL PORTLAND INC = GENERAL REFRACTORIES = GENERAL STL INDS INC GENERAL TIRE RUBR 354z = 37'737910 355 = 3793521C 356 = 37956810 = GLEN ALDEN CORP = GLOBAL ARINE INC = GLOBE UN INC 357 = 358 = 359 = 360 = 3813171C = 5OLF 382338'10 = GOODRICH 3827481C 3834921^ = GORCON JEWELRY = GOULD INC 361 = 38515410 362 = 38741210 363 = 3874781 364 = 3P92801C 365 366 367 368 369 = = = = = 3906041C 39109010 3'14421C 3915141t 3923881C 370 371 372 373 374 375 376 377 378 = = = = = = = = = 3q,304610 4013701C 4020641C 4024961C 4027841C 4042451C 4058911C 40609010 4062161C = 4102521C EN' WEST FINL COR B F CO CORP = GRANBY MNG LTD MGMT SVCS IN = GRANITEVILLE CO = GRAY DRUG STORES INC = GREAT LAKES DREDGE = SRE-AT NORTHN NEKCOSa = GREAT W ESTN FINL COR = ,GRANITE = GREAT WE STN UTD CCRP = GREATER WASH INVS IN = = = = = = = = = 379 = 4083061 GREEN GIANT CO GUARDIAN INS CORP GULF WESTN INDS IN GULF RES CHEM CORP GULTON H M W INDS INC INDS INC HALL FRANK B & CO IN HALL F PRTG CO HALLIBURTON O HAMVERMILL PAPER CO 382 = 41C3421C = ia1NDLEAtAN CC DEL HARS.AN = HANDY = HANES CORP 383 = = HAROURT 384 385 = 41334213 = 4136151 = = HAR ISCHFEGER CORP 390 381 = 4103061C C L HOST CORP 41163 1 1C A ri ? HS BR4CE JOVAN y FI C c rl_ C 1_ MU C C CO'VE 3R6 = 413g 7510 = 'ARqPIS 387 38, = 4161941C = 4174041C = rlA- T E = HAR TZ 389 390 391 392 393 394 395 396 397 = = = = 4198661C 42159610 42268620 4228341C = = HAZFLTINECRP = HEFS INC = HEILE.t^N G CREWINS = = 4230741C 4,2323610 = HEINZ H J CC = HELENE CURTIS 398 = 4281821C = 42343410 = 4234521C = 42786610 = C I INDS I = = 4344341C = 43508110 = 4360o210 412 COfUUNIC CCRP 4422721C = 401 402 4C3 = = = = = 409 = 410 = 411 = TN DEL 4376141 4381283C 44050610 4410611C 4410b510 44197410 4415601C 44175 81 4C0 404 405 4C6 407 40 CCRP HANKS = HEL E PRCCS INC = HELMERICH1 & PAYNE IN = HERSHEY FOCS CORP = HEUBLEIN INC = HIGH VOLTAGE ENGR 0 = HILTON HOTELS CORP = HOFFYMANELECTRS CORP = HOLIDAY INNS INC = HOLLY SUGAR CO'P = HOMESTAKE MNG O = HiON1CA MOTOR LTD = HORN HARDART CC = HOSPITAL AFFILIATES = HOSPITAL CORP AMER = HOST INTL INC = HOUSHTON MIFFLIN CO = HOUSE FABRICS INC 399 = 4298121C C A LISTOr 432341C HOUSTON AT GAS CORP 413 = 44228110 = HOUSTON CIL MINERP, 414 = 4426721C = HOWARD JHNSON CO 415 = 44485910 = tHUMt.NA INC 416 = 4455823C = HUNT PHILIP A CHEM C C C - . C. 417 = 4480961C = HUSKY OIL 418 = 44e4991C = ilUTTON 419 = 420 = 4497441C = 421 = 422 423 424 = 4t515421 = 451h5301 = 4527221C = ID .5C = IDEAL 425 426 = 45303820 = 4547581C 427 = 4562991C 428 = 429 430 431 432 433 = 45732630 = 4576411C = 45765910 = 45768610 = 45810 11 0 434 435 4566231C f. I N A CORP pWp CC BASI INDS TOY CRP IMPERIAL CIL LTD INDIANA GAS INC = INDUSTRIAL NATL CORP = TNEXCO OIL CO = INL ND CCNTAINER COR = INMONT CPP = 4587021C = = 4523001 = INTERLAKE INC INtT:RATI CNAL = INTE T I C N AL I 'NTER'! T I CN L = 438 = 4596501C 439 = 440 = 4600201C 4595731C 459R,841C INC = = = = = 437 INC = IDEAL = IMPERIAL CORP AMER 436 = 4595061 C ROUP HYDROMET4LS INC 4490061 45138910 LTD E F INSILCO CORP INSPIRATION CONS COP INTEGON CGRP = INTENATICNAL = INT-R0NAT IC"-'AL = INTER.NATIONAL cUSINE FLAV OR HAR VE S HLD GS 3 ' I IERA 4N G CO P SAT I C'' NAL r cN T S FIL E: UL L LI ST" 441 = 46325412 = = 4 609' 4L3 = 4607541C 444 = 611431 445 = 4614.810 446 = 4622771 448 = 4656321C 449 = 4656431C 450 = 47024510 C. 451 = 4710161C 452 = 453 = 47836610 C C C C. C .. C C, C NAL RECTIF 4710701 456 = 4810831C 457 = C INT-STATE = INT-RWAY CP = I.V BRANS S TOR S CC CIVERSIFIE BE- F PROC SSORS IPCC HOSP SUDPLY COP = ITEk CORP = ITEL CORP C IN = JAMES FRED S = JANTZEN INC = JAPAN FD INC = JOHNSON CTLS INC = JONES LUGHLIN STL = JORSENSEN = JOSTENS = K 48251610 458 = 483C08 31 = 4830441 CO = = 454 = 48003410 455 = 4808271C 459 IN T-RNAT I I OW = 4626141C C C _ C' 1C = INTERPUBLIC lRCUP 442 447 A L M ROYAL = KAISER C EARLE CC M INC DUTC Ar ALUM r CHEM C KAISER CFEM M ; SYPSU 460 = 43306210 = KAISER INDS CORP 461 = 4309 810 = KAISER STEEL CORP 462 = 4840981 = KANE MILLER CORP 463 = 4851701C = KANSJAS CITY SOUTHN I 464 = 48602 1 C = KATY INDS INC 465 = 48638610 = KAECKI ERYLCO INDS 466 = 48731410 467 468 = 48765610 = 4880441C 469 = 48917010 470 = 4927462C 471 = 4934221C = KEENE rCORP INDS INC = KELLR = KELLrGOOD CC = KENNMETAL INC = KEWANEE INCS INC = KEYSTONE CONS 472 = 4937821C = KIDDE WALTER 479 = 51 INDS I CC IN 473 - 49436313 = KIMEPRLY CLARK CORP 474 = 495;901C = KINGS DEPT STORES IN CO 475 = 4976561 = KIRCH C 476 = 50017013 = KOE FRIRI IN.C 477 = 5006G21C = KOPPERS 478 = 50062 01 = KORACORP INDS INC DE 261 = KROEHLER 480 = 503185810 = L F E CORP 481 = 50221010 T V CORP 482 = 5024441 = L = L V 0 CORP 483 484 = LACLEDE = LAMSON = ;5055 881 = 5136961C MFG CO GAS CO SESSIONS CO 485 = 51839010 = LATROBE STL CO C C, 486 = 5218941C 487 4R8 483 = 92206 1C = LEA SEWAY TRANSN CORP NORTHRUP CO = 5241921C = LEEDS Pc = 5248581C = LEHIGH PCRTL AND CEM' = LEHMAN CCRP = 5251741 490 491 = 5262641C = LEAR SIEGLER INC = LE JOX 404 .INC CC = LEV I STRAUSS = 5274 802C = LEVITZ FURNITURE CrOP = 53'O31Jl r = LIEP cY OIENS FORD CO 495 = 492 493 = 5273641C 530 331C = LIBBY MC NEILL LIB E cVC"C T I C ITT ? FILE: LI STDR CULV'JL 496 = 497 = 53625713 498 = 49 = 5398211C 5230 = 53037 5 3;2 1 11 544241C 501 = 54219513 C 502 533 = 54229010 = 5435'1C 504 = 54777913 505 C C C = = 507 = 508 = 509 = 5 10 = 511 = 512 = 513 = 514 = 515 = 516 = 517 = 518 = 5C6 C, C: (.,. 1 C ;--1 ·U. I E ...- _;.. C 1 0 ìiL·· 3:. -·-- C (. S C [IC CC2L CORP = LOCKHEED = LOEWS = LODGNTgIN CRP = LONE STAR INDS INC = LOR AL CORP = LOWENqST IN SCNS = LuBRIZOL CCRP = = = = = = = = = = = = = LUKENS STL CO 520 = 521 522 = 56631 91 = 5664721C 523 524 = 5697131C = 5711541 56532110 LYKES CORP ·, C M E A I INC CORP * S L INDS INC ANDREWS £ FORBES ,' . M;AC CONAL MACKE CO E MAC MILLAN INC 4MA-Y R H AA ISOOI FD CO0 INC INC F CO MAGIC CHEF INC MANHiMATTAN INDS INC = MANPOWER I NC = 'AR ATHON MFG = ,4MARCOR INC = 'AP EONT CO CRP = MAR ION LABS = MARLEY CC INC 525 = 57144310C 526 527 528 529 = = = = 57335310 5743551C 5752161C 57974610 = = = = MARTIN PROCESSING 53C = 58003310 = "'C DERMOTT MARQUETTE CC IN MARYLAND CUP CORP MASSEY FERGUSOJ LTD MC CORD CORP J RAY C = MC GRAW ILL INC 532 = 58123810C = C !NTYRE MINES LTD 533 = 5813210 = MC iKEE CCRP 534 = 58210310 = M1C LEAN TRUCKING C3 535 = 5322731C = MC LOLTH STL CORP 531 = 5~936451C 536 = 537 = 58507210 = MEDUSA 538 = = 539 540 = 586051C = 59644710C 58283410 58574510 541 = 58943310 542 = 59018810 544 = 59082510 545 = 59160510 546 = 5916931 547 548 549 550 7 = LIs RTY -0RP = LIC;;EL CRP = LITTCON I NDS 5496610 5508901C 5526531C 5527101C 55371310 5542051C 55430713 5545283C 5547901C 5561391 5574 8 10 55910810C 56287610 543 = 5906551C ·' *- CCtNVE- AT ICN, L VtC ITO , I; 54927110 519 = 5641811C K. & = 5945031 = 5947291C = 595 531 = 59515210 = MEiAD CORP CCRP MELVILLE CRP = MEMOREX CRP = MENA!SCO FG CO = "EREDI TH CRP = MERqILL LYNCH & CO I = MESA PETE CC = MES TA ACH CO M ET-RO GOLDwYN M4YER = MET'OED I A INC = ,NTICVLIGAN AS UILS C = ICF-IOGAN SUGAR CO = "'IC CO0T INC = 'I COWAVE r ASSOC I NC = r 551 = "I" = 59539010 CCV-JRSATI[NAL A LI STPR ULU..'~UL FILE: TEL CO'NTINENT o 552 = 52583210 = MIDSLE SCUTH UTILS I C 554 = - MICLAND ROSS CORP = '-37711C INC LAES 9MILES = = 5992921. 555 = 60175310 = "ILTON BRACLEY 556 = 60411010 = MINNESOTA 557 = 60598013 = MISSION = 636191 1 = MISSOURI PC 559 = 606193C = ;TISSOURI PAC RR CO 560 = 553 '5rP - C ·. C; 6 3301 r = PR C S LT C ROUP IN INS CORP 0MOHASCO CCRP 561 = 6381831C = MOIHAWK DATA SCIENCES 562 = 60 830210 = MOHAWK RUBR CO 563 564 = 60872720 = 6D87441C = 565 = 566 567 = = 568 569 = 5 570 = = MOLYBDENITE CORP CDA = 'OLYCORP INC ,ONARCH MACH TOOL CO - = 60-91501 = MONCOGRAM,INCS INC DE 6097621C = MOESANTO CO 6116621 RES = f'CO° E MC COPMACK 6157981C INC OR'RISONh KNUDSEN = 618441C 6190 7 51 C = .'OP.SE SSHOE INJC 571 = 61935610 = MORTONNCRWICH PRODS 572 573 574 = "UNFORD INr = 6261441 MURPHY OIL CORP = 62671712 = = 6271511C = MURRAY OhIO MFG CO = MYERS L E CC = N C R CORP 575 576 = = 577 = = 62915 6 10 : 578 579 58C = 6294491C = 62985310 = 63085410 = 6312261C 581 C52= 62q4542C 62886210 63243110 = = = = = INC L INS N V F CO 'IALCO CHEM CO NARCO SCIENTIFIC INC N4ASFUA CRP NATIOCNAL AIRLS INC 5R3 = 63256710 = NATIONAL AVIATION 5P4 = 6348921C = NATIONAL BELLAS HESS 585 = 587 = 6361801C 63512 312 CAN CORP = 'ATIONAL 586 = 6354171C = 'IAT ITONAL CITY LINES = *4 T IONAL GAS CO FUEL 588 59 50C 591 = = = = = NATIONAL GE CRP 6362141C 6363151 0 = 4NATIONAL GYPSUM CO 63641810 = N'ATIONAL HCmES CORP 63648610 = NATIONAL INDS INC 5032 = 6368821C = NATIONAL MEC CARE IN 593 = 6368861C = NATIONAL MEC ENTERPR 594 595 = 6372151C = 63774210 = N'4TIONAL PRESTO INDS = 'IATIONAL STD CO CH 596 = 6377761C = NATIONAL STARCH 57 598 = 6378441C = 6380971C = 'NATIONAL STL CORP = 'JATIONAL TEA CO 599 = 63835210 = NATIONL o_ C C 'NATCMAS CC = 'EI SNERS 8ROS 6C0 631 = 602 = 64074510 = 603 = = 'IE\VAD 604 605 = 64405212 = 6482110 = 63876C1C 64021 1C 6414231 UN ELEC COR EP TUNE INlC INTL CRP PwP NGLAND = NE = NEW PROCESS Sn A3 CO C EL ONITD>: S FILE: CULMUL 6C6 LI STmR = 65C I1 1 607 = 60O = 6 1C9 = 6514271C 65163 91 6535221C 610 = 6535561_ 611 = 65638910 C 612 613 614 615 = 617 618 619 620 621 622 623 624 = 6686051C = 6687071 0 = 6703461C = 67140010 = 6745991C = 6763461 616 C C C C 625 6 26 627 628 629 C 630 .. Cr`: ( ( C 631 632 633 634 635 636 637 638 639 640 641 642 643 644 C ;r·'":·G I·r C ·' .i··'""; c·;--- U .--o:·ir.....·. *i·-·r : tr C I ·· i_ :;F `'i:···) -·i 0 --i · ··,_ · ;t·· "'. _rc .. 645 646 647 648 649 650 65678010 = 6570451C = 6584081C = 66550010 = 66752810 = = 679043 1 6 01(6652) = 6820631C = 6825051C = 68900210 = 690,92010 = = = = 6901051C 6933631 6907341C 69076810 = 69149 73 C = 6935061C = 6947501C = 69562910 = 6964801C = = = = 69764310 6980571C 69846510 6988221C = 7111110 = 7025441C = 70905110 = 7099031 = 7113211C = 71110610 = 71344810 = 7140411C = 71602610, 651 = 71645110 653 = 71654410 = 7165491C 652 JEW ,t i 4 YORK TI ES CC HF LL LC .S FARM ING ONT .'RA 'NG CO2P OHAWK PWR IAC IAR.A SH CRP S INC 'ORISRTS NORTH MERN COAL COR ', ORTH AMERN PHIL IPS NJORTH CENT AIRLS INC NORTHERN NAT GAS CO NORTHWEST INDS INC ,NORTON CC NORTON SIMCN INC NUCOR CORP OAK INDS INC OCCIDENTAL PETE OGDEN CORP CKLAHOMA NAT CORP GAS CO CLIN CORP OM!RK INDS INC ONE IDA LTD OTIS ELEVATCR CO OUTBOARD MARINE OUTLET CO CORP OVERSEAS SHIPHOLDING OWENS CORNING FIBERG OWENS ILL INC OXFORD INDS INC P P -S INDS INC PACIFIC PETES LTD PAINE WERBER IC PALM BEACH CO PAMIDA INC PAN AMERN WRLD AWYS DANHNDLE EASTN PIPE PAP=RCRAFT CORP PARKER PEN CO PASCO IlNC PE'N'YLV PENNZOIL NIA CC DEOPLES DRUG STORES PEOPLES PEPSICO GAS CO INC PERKIN ELMER CORP PETER PAUL INC PETRO LEWIS CORP PETPOLANE INC PETOLEU v RES CRP 6'54 = 71 31671C PHILIP MORRIS INC 655 = 71 8321C PHILIPS PHOENIX 657 658 659 660 = 71915110 = 719,651C = 7201!of1C = 72447910 = 725106 10 656 L PWR INDS INC STL CORP PICKWICK INTL INC ?IEDCYONT NAT GAS PIT'Y BES DITTSBURGH INC INC FORGINGS ~ ~ ;c-ki~4 I~ r FILE: CLJLAU LLI S T PITTSTON 665 ~'OL 'R ID CRD PC'nDEROSA SYS INC 666 667 C C C. = 7324361 = 73620210 = 73762 81 74035121C 7431071C 74449910 671 = 74463510 672 = 74533210 673 = 74579110 PREPtIER IL PRODUCTS RSH 674 675 = 74740210C = 7497510 676 677 = = 678 67 = = 74 73 810 75215q 10 7545861C 7547221C = 75764010 C: C. S-iL :· .:-·· idr· A CORP T E CORP RANCO INC RAY`]ESTOS TOOL "=EED REICHOLD REL IANCE 685 = = = 7594661C 76033542' t L I ANCE EE = 76088110 = 7613391'C 691 = 76140610 692 = 76152510 693 = 7616861 604 695 = = 606 = 7631211,0 C 76168813 7617631C 7631721C 6q8 = 7664811669 9 = 76775410 = 701 702 703 = = 704 = 705 706 = = 77379410 77571110 77633510 7766781C C 713 = C 711 712 713 714 715 = = = = = 1C 707 78024010 708 = 789029110 CC ROS INC CHEMS INC ELEC CO ORCUP REDLLIC INC CCRP REPUBLIC FINL SVCS I REPUBLI STL CORP E,EARCH COTTRELL REVCO D S INC REVERE CCPPER REVLON I NC REXtHA, CCRP IN ; BRAS REX NORD INC REYNOLDS ETALS CO RIC = 77055310 7707061C MANHATTAN RAYP'OND INTL INC .ED.... INDS INC 75945710 7810881C 7812071 78224220 7835491 783.781C 784 311I 78462310 ( C R 7592001C = C R = 689 INDS INC INC = 687 7604091 688 = 7607791C CHEM N' MEX QUAKER OATS CO 684 709 nl PULLMAN 683 776806 L·" PUBLICKER CO PUGET SOUND PWR C LT REEVES 686 CORP SVC 75855610 = = C ;·':;·· 26 01C PUBLIC = 697 C PCOPTEC I C POT LATCH CORP = = 690 C ENTERPRISES = 682 C L YBOY 669 670 6P81 = 75 C CO PITTVAY tCRP 668 680 C C'V RS4ATIrCJ'J.A L MC IT,i 661 662 = 725701 1 = 72578610 66h3 = 72811 71C 664 = 731 951 C A ARSDSCN C ICHARDSCN RIEEL ERRELL TEXTILE CORP RITE AID CP ROn3ERTSON 1ROB,INS A H H H r INC ROCKOWER BRCS INC ROLLINS INC RONSON CCRD ROPER CORP ROSARIO RES CORP ROYAL CRCWN COLA CO ROYAL RUBZERA " UCKER NCS INC ID INC CO TOGS TNC RUSS RYDE P SYS INC S A A S C 0 CORP S CONS SVCS INC INC I FI L C : C L;L 'UL 716 = 7867S9 7 17 713 719 = 79744 1C = 799q5010 = 8006811C L I 720 = 30202 1f 721 = 8020371' = 8037011C = 724 = 80461471C 725 726 = 8051751C = 80685710 727 728 729 = 80912310 = 8091801C = 8097411 = SAV ON DRUGS INC = SAVIN USINESS A14CHS = SCHLUMBERGER LTD = SCOA INDS INC = SCO3 LAD FCCOODS INC = SCOTT FORESM-aN ; CO 730 = 81062310 = 731 732 733 734 735 = = = = = 8113692C = 8115171C = 8116411C = 81196213 = 8120991C = SE- C 736 = 3125571C SEATRAIN C 738 C STCP INC SCOTTYSIC CONTAINERS INC SEAP.OARD CCA4T LINE SE4_CARC WORLDAIRLS SEa GRAVE CORP SEALECTRC CORP LINES = 8357161C = S00 = 84129710 = 757 758 = 84316310 = 8434561C = SOUTHERN INC GAS & E = SOUTHERN NAT RES INC C 759 = 84367310 = SOUTHERN C, 760 761 762 = 8440281r = 84723510 = 8475411C = SCUTHERN UNION CO = SP RTON CORP = SPECTOR INDS INC = SPRAGUE LEC CO = SPR INGS M'LS TIC = STA LEY A E MFG C O 746 747 748 749 750 751 :...;... 752 C ·7. C ····it "- CG C = = 8177151C 763 = 84933910 i ;c 4 756 C ' S4V 755 740 741 742 743 744 745 (2 '" = 753 754 K .. 1. ·-:T 80460310 INC = Sq VOMATION CORP = 81913910 = SHAKESPEARE CO = 8194701C = SHAPELL INDS INC = 8208631C = SHE AR SON HAYCEN STON = 8226351C = SHELL OIL CC = 8227372C = SHELLER GLBCRE CORP = 82434813 -= SHERWIN ILLIAMS CO = 8261832C = SIEGEL HENRY I INC = 82641410 = SIEPRA PAC INDS = 82662210 = SIGNAL CCS INC = 82867510 = SIMPONDS PRECISICN P = 8287631 = SIMON &. SCHUSTER INC = 8301641^ = SKACGGS COS INC = 83064310 = SKI L CORP = 33083310 = SKYLINE CORP = 3211010C = SMITH INTL INC = 83269613 = SMUCKER J M CO = 83408610 = SOLA BASIC INIDS INC 737 739 ,.... -·: :-I ··il.,.·::· .', SAGa COR' = SANI CIEGO GAS ELEC = SAJ ERS aS.SrC INC = SAN-JSI'MO E LEC CO = SAWTA F IN CS INC = SA!N TA FE I"TL CORP = SA.GE:.NT w~EL-H SCINT 723 . ;.i.;.. rCiVERAT ION'L Ct IT r 722 C i TPDR LINE RR CO SOUTHDOWN INC RY CO 764 765 = 85178310 = 95256310 766 = 767 = 3538731C = STANDARD PRUDENTIAL 768 760 = = = = 770 = 8538191C 85410610 85423110 546161C = STANDARD PRESSED STL STANDARD SHS INC STANDEX INTL CORP = ST. NLE Y ^KS -liFILE: 771 C C C CULY',JL = 86130410 = 8615721 776 -= 36316310 = 791 = 86 901 92 69 71 61 8716161C 7'264 91 = 8746871C 8753821C = 7q2 = 8760431C C. C. 796 = 8791311' 797 = 8791991C 798 = 87933510 799 = 87986210 800 - 8 1 60 9qlC 801 = 8816941C 802 = 88249110 = 8825081C = 8892931 805 = 88261031 806 = 83289510 8C7 = 884Lj 2 1C 808 = 8847531C 809 = 88634B1C 810 = 88 642310 C C, IC CAMP I""C & EBSTER = INC SUN INC = SUNDSTRAND CORP VALU STORES IN GEN COR = SURVEYOR FD INC EL = SWNK IN = SYNTEX CRP =T R W I NC = TALLEY IDS INC = TAMPA ELEC = TANDY CORP CO = TAPPAN CO = TEK TRONIX INC = TELAUTOGRAPH CORP = TELEDYNE INC = TEMPLE = TESORO = = = = = TEXACO TEX S TEXAS TEXAS TEXAS INDCS INC PETE CORP INC INDS INC INSTRS INC OIL SAS CORP PAC LC TR = TEXFI I CS IC = THICKOL CCP = THOPSO J ALTER C = TICCR = TIDEWATER INC 811 = 89673512 812 = 8872241C = TIGER INTL = TIME INC 813 = TIMES MIRROR CO = 8873601C 814 = 83917510 815 = 8902781C O g CO 793 = 878 5 1 2 1 C = TECFNICARE CORP 794 = 87852110 = TECItNICOLOR INC 795 = 8785421C = TECHNICOh CCRP 803 C. STONE P VAN 3= 6844313 = 2UPER\ARKETS 790 = 87512710 C, = STCOJL = 8680351C = SUPER 7.86 = 787 788 = 789 C T VE'. S J = 86158912 = STONE CONTAINER CORP 781 = 8667621C 782 = 86732310 7 85 C rC'CV-mS&ATTC.6AL 777 = R,2.9712 = STOP S SHCP COS INC 778 = 863%6310 = TUC3B KER ORTH INGT 779 = 8666451C = SUl 2HE M CCRP 780 = 86671312 = SUN ELEC CORP 783 C A = 85772110 = ST 1FF CfHEl" CO = 8583751 C = STEEL 0 CA LTD 772 773 774 775 784 C LIST RQ 816 = 8905161C 817 818 819 320 821 891508 = = P.9289 21C = 3933491 = 8934851C = 89355313 = TCLEDO ECISCN = ORP TONI<A CO TOOTSIE ROLL INCS IN 1C = TOTAL 234810 INC PETE = TRr -COR INC = TIRANE CO = TRA'S WORLD = TrPASAMERICA NORTH AME AIRLS CORP =TRSCO1 LINES =TANSOH-I C FINL -CP 823 = 8938471 824 = 394l151f = TR SWAY INTL CORP 825 = 8954351C = T2I CONTL CCRP IN TrIT SY -1 13FILE: CU!JLUL LI"T 826 = 8o5$611C C C ., C C. C C ( .·- C. i-L -'_ j ·· ··. ·-· i .··1.--. 827 = P958951C 828 = 8965221C = T INITY 829 33 831 = = TROPICANA = 9012211C 832 = 9921201C Q970931 = 8.3831C 333 = 9321 210 834 = 90255010 835 836 837 838 839 840 841 842 = = = = = = = = 843 = 90791g1C 90268610 9028731C 9032001C 9034221C 90344310 90427410 90553 1 9055811C 844 = 9093131" 845 = 9096631 -,.. 848 849 85F 851 852 853 = 9134849C = 91056210 = 91063710 = 91067110 = 9106810C = 9112131 854 355 856 857 = = = = 91135813 91153610 911 82 51 91184210 TUC SON S I C PRODS INC I 6; ELEC 43S = TWENTIETH = TYC LA S = TYLR I"'C CE'TY INC CORP = U = = = = = = = U G I CORP U M C INDS = = UNION CARBICE CORP UNICN PAr CRP A L INC U O P U V tINDS INC INC INC /ARCO INC DEL UN qCO INDS INC EL UNION CAMP CORP = UN ITED AIRCR AFT PROC = UNI TED BRANCS CO = UNITED CORP = UNITED = UNITED = UIT ED = UNITED FINL CORP CAL GAS CORP STY CORP ILLUM CO = UNJI TED I NL CORP = UNITED INNS INC = UNITED NUCLEAR CORP = = = = UNITED REFNG CO UNI TED STS P FGN SEC UN I TED STS FID GTY UNITED STS FILTER CO STS GYPSUM CO STS LEASING I STS SHOE CORP STS STL CCRP STS TOB CC CORP .JSL IFE CORP = 865 866 = 91819510 = 91 31410 = V C = V S 867 868 869 870 = = = = = VARIAN ASSC INC = VARO INC = VIACOM INTL INC = VULCAN MA TLS CO 9222041G 92227210 92552610 92916010 874 875 = '3316910 = 93369610 C, 876 877 e78 879 90 = = = = = 93405110 1C 93443610 93833710 9396491 93440 CO FOX = 91731-31C C. K IS PAC CRP 864 0 n = RI A"IGLE TR!N"lGL' 858 = 21202710 = UN I TED 859 = 9121231C = UNITED 863 = Q126051C = UNITED 861 = 91265610 = UNITED 862 = 9127751C = UNITED 863 = 9133531C = UNIVAR 871 = 92 97691 872 = 9297941C 873 = 9323551C 4 T C 'Va = = 846 = 9101110 847 = 9103141^ -CC A A CORP r CORP = WAC HOVI4 CCRP = WACKENHUT CORP = WALLACE 'URRAYCORP = WALTER JIt CORP = WANO LAOS INC = WARC FOODS INC = WA!>NER f SWASEY CO = WARNER C0NMUNICATIrnN = WASHTNGTOCN GAS LT CC = WASFIi4GGTTCN POST CO r T !IT NL M Cr!, I T , F I-E : CULNIUL L IST a. -C' 'Va-RSTTTCNAL ( = ;3 R 2 , .. C C 3 883 = = 3 85 = 3 86 = 887 = 888 = 889 99C = ..'. ...--.- I- 8391 C i- 6199093 iJ·::: 1_ = 9758761C 904 = 97738510 9 C5 = 9774801C C (? 0 6 = 9790971C C 9 07 = 9781651C 908 = 98006510 909 = 9808811 C ,r ··3 C i .·-.·; C -·a?----··· p, 3:'-1; . C, n- i 1 ·--- ;·- ;· "c; 'p · 9 15 9 17 r r;-'·· · 9 12 9 16 -,--·?·,;,FI-··r· ?. 'Y`.'- 899 = 9640601r 900 = 96668010 it: --·IC· = 98413 8C 13 = 9388571C 9 14 = 9890701C r! :tl··; ·, c C- C IN CO NORTH AE ESTERN PAC INDS IC = WESTERN PUFG INC = WESTVACO CC9P = WESTMORELANC INC WEYERHAEUSER : CO WHEELABRATCR FRYE = WHEELING PITTSBURGH 03 C. = AIR LINES 897 = 9631531C 91 0 = 9825941C 9 11 = 19412110 '1 LTD I'iC = WE THERHEA C rO = WEB£ DEL E CORP = IWESCO F INL r3RP = ;WEST POINT PEPPERELL = WES TCO ST TR ANSM I SS I = : ·-· .,· 'AN CRP 396 = 9 62892 = 9574431C = 9746371C C = = 9621661C 98 = 96362610 :·· cilr:t- ASHINGT CN STL '=4A TE Mv 'T IC 9575861C = ESTERN = WESTERN 95804310 = 9590901C 8 94 = C 94701 1 9471511, 9474231C 9508171C 9554651C 95751810 892 = 95926550 893 = 9615481C 895 = = 941 631 8 I 40144 1C = 93939910 = 9 95691C = 989241C = WHITE CONS INDS IN INC = WHITE MTR CCRP = WHITTAKER CCRP = wICI ES CC P = = = = WINNEBAGO INDS INC WINTER JACK INC WITCO CHEP CORP WITTER DEAN ORGANIZA = WOLVERINE WCRLD WIDE = WOMETCO ENTERPRISES = wOODS CORP = WOOLWORTH = WURLITZER = XERCX = XTRA F W CO CC CORP CORP = ZALE CORP = ZAPATA CCRP = ZENITH R IC CORP = ZIMER CORP = ZURN HES INDS INC MCC -115SAMPLE PORTFOLIO -3 J-Ji 4 ~-I.~ C: ~ .. 4.' - 4-4,4 0,, J = L1 4-4 -4 "--.,-. -(9 -4 " -4-4=~~3i -'3 4-4 ,,,.,.,-- , t I 3 4Z . .- 3 lz-; 3- -- n· -=.1- r4-4 ' 43 o 3 I 4-.3. -- 3 3 C23 - N INN =3 zr- 3 ~ 3 3~3330333333 33 4- 3 N33 3 3 333333333N3'3333 3 ; 3 33 4'- 33~3' 3 3 4 '3 ' 3'3 23 3 3 :1 C I "I '34.340 f~~ ~~3f~~~~f~~c~~g~'-- X. CJ~~c~r~cO~~'3-4 -4 .Z ~ ~~O ~~r~~~~~~~~=9-4 0'' D~~~'Ol" .44- - -4i 4 -441jF-~ -)= ')4 .4 = E-. V)Irc F- 1.'4 - -'r 3(7-r -4 E- 41 ci 7 C- -4 '7 ~' .33 ~ ~ ~ 3 '3 N 3C '-3. 143CC3 3 ; 0 '3 - 3 3 3 3"' i 0333 IN; 34 '3~r 1 '3 C Z 3 3i := :-I 13 = 3 = - z~~~3 I = 340 I I '33 - 3~ I-4 4'-L ."' .3-4 ,4 ,' 7 3133 S~~33433' ..-- -3 '7(7 I I 33 4 °~ 334' -3- '2,= =0 Nj 03 Ll o,",- 33 I.. . F-L 7j2 3/ 34j-'43. ~3'33z3473 3 3 3 (N 3 N333'3u '- 3 _ 3 3 ('4, '3 '3-'333 - 3- .3 F3-I 3 h - jC` ·i e t t IIi I I i t -3 'i : ,z--. -7 I- -4,41ri i; 4" : 0' ) - 1-40L "--F---. ,: i j .J '3 3 C -41-'--'-4 33 t·-l-·J =·= L (7 - if. I C- 5 I I I 4 I Ii I4 c- ,CC' ~-, ,: Ir yj - _1 -' ~ " , r'; 'U2 f',- I c f I - L.% I _-~ I . - P-- - "C' i C` f LT i I -' -~~." 4-:~_ i- -- ~ -' ~ -- ., ( 0 '3 - '4 3 -3 'C' 44-' '~ 3- - ' C 3 . C C 3 I - LI7 L,~~L - " . 3F 3 r'3C 12 3 -. ~ '3'~ , _. - -=-, ~ ~~ ~ ~ 0.~ ;~' ~ ~ ~ C 4334"4 4I' J:"- ' "'--~ ~'" -"--- 3 - ,'TN; "' ~ .33-4 4 4 I 47333 N 3 ('.4-4' 3 - 3'4147 I 4444III 4 II I 3 ~-C 3 '~ *- ' -4 '- 3333 -dW k.d- . .3...1i I~ . II -116- Appendix II COMPUTER PROGRAMS -117PROGRAM 1 EXEC FILE 0 0 rz~~~~~~C o o o 0o 00 o o000oo rr c o (N 7( ab= - C > C c = U . C. U -- r. c. : C C~ E U C - _- - - I~~~~~~~~ __ MI -1181 PROGRAM a 0 -: C C. O C C Cocooocoo0oO C, C C O cc 0 C IDC000 O O O :> °O O O O O O ° O O O O O O ° O OC OO c, 0 C) c C af" C O O O' a 0 0 r : c C-D4C 0 C, 0 C, CD 0 D 0c C:, 000 U LN CD C) a O O O 0 O0 0 0 0000000000000000000000000000000000000000000000OOOOOOCCCo _ CDC, C O O OC, o:O OC C:: O C* O '~ C, : . > O CD OOOO O O O O O O O O O O O ) r -0 0 O 0 C, O0 0 YtCL 00000000 C, 00 O O O O OC _ 0v 0 O0 a O C oOoocooODOO 0 O °O O O O O O OOO OCG'O O OO C O OO OO O O OoOo 0 O o OI In C: 0 000 O O OO e C, C O O o O C' O O COOO O O C.OO OO O C C C OO C G C r7 ooo§§ooooog oCo O O t -4C,0 C L CD 0 0000000,0:C0 00 O 0 0 O CO OO CO o O o O C, 0 e 00 L O O O O o C, I C 0 O C -5 0 C Cr 0 W "O C C~ ~O 0 O COC O O-D-O O 6Co O "r' t C ~ "% at ~~5H 5~~~~~~~~_ C: .r: _ ~~ ~ Sr., C ; C [..- '" c'3 f~ ~~~ " F0: t., C Y0 - = A _ C 2 E,,- · <- c . H E- _ F- C 0~~~ C H - - ~ ~ ~ _ FC - _- F- .C O o _~ _. CC H 0 0.- H _ C .C- F N F- -C C -C C-H CO: C F -. _ C: c-'~ C C- N CC _ - rE- C 3r FV L- 11 0 - -i _ r 0 C* C CF 00_ . C : - =p.2L~ 0 _- C c-.z: . ^C C":C * t C, CC- _ 'C_ F-"C _ F02 _ C 'C -o X 00 F F- _ = CC F 0 F-C H F" 0' _r ~_ FH 0 C2 0 0 C . C_ ; . -C t C i- . " C_' .) F--; F- F- [COL F-~~~~~~~~~~~ _ C - _--~ ''-- F-'-C -s F C0 E-- C_ _' C C-C 00- :1 [O C_0 _-~ Cos 'ICs - C F l.C 0-. ! = C C.- E- 0 )C [~. -_ r 0 C" - 1C CC - -r, C_~ C oCO. ., G _2E1- H2Z C F- IS[-I' F- C cC- ¢ r ~FF-C- C'v C, C, C: _ - - ,.- ',. o _ .. 1 'C C '- C __ c '_I L) _ I 0, -118. I n C: *_ . C, C COC OC 0000000000OO00 0 0 0 0 0 0 0 0 a C :, C O O O~111CD a IL,rc,CF _ rO 1:1n L-)L' "I 0 ZD >0 D If U; Z 0 011- 0 I 000 oooooo ooo, rz o C.^ C ooooooo 10oo -o CD C o CD o 0 0 00 C' 00 0; 00 co 000 O C C O C 00 0 0 0 0 C O o : r O O C,-. 0 C z r- = G C, Z0000= 0 IC I r 000 r-N - r N 000 C o 0000 o o- 0000 o CD C Oo o a:o a: oO000 o o o 000 (,o oooC0c 00 O0 000 00 000O0O0 0O00O000000O C o C) C C00 CD0 gco ooooogooggg ~ O 000 0 O 00000 000 00000 O O. I- CO C C 0 c, O0 0 a0 0 0 0 00 0000000 000 r o C CD - 0 0500000000000 Do 0. C 0 0 3' 0 C' 0000O000O.-.o ' 0 CC 0' 0 0 - oo CO 00000O 000O0 Ca 0 0C.) CO CC C )O c2~_,, t C0'000CDOCC C r: _ In C r CZ C F- 0 = H E- HInO F- ,,c o u C E: O F- C - °n zz C · "C Cjc C C - C :":. ,-1 ,. C Es C C ~O 0: [-. C~ umo Er. Q In n0 OH H) 1 u ',C C: 0 H ; :0 . ._ 0. EL' C 3 C I Lr - C0. C C -cC 0 ~, C, r. :. ,.C- v_, C- *' ( gN.- Or_cE: C ' I' C c, : ' 000 C - :. ~' -0 F- __O Co C. _ F :0 - * Y F- . 0 =, m: C X C: c-. .0 _ : H 0 F- _ g,, OH o H r E H; 0 00z- -C E- l C C * ^ 0 CC,c3 C::L =' '-'¢ C . r4Ot Cc o-. OC H = .r -- tt~.- -,a; 0 Lu C -; z C E- NC 0 iE u OC :~ OF _ o' [_ ~cr 0 u 0 Cs F IPON ," FHO 54. -II H 0 -_ e- _- _.,. - ' O 0.0EC > 0z ,c _ ' .-r I _-- U-- ~1. L- -~ H~C I C 20 C . *^ . C -i Q L: C- '. b .- Ln r I C 0 C t0 C C - - ^ I C O- N _, I !NLN~- 0.u'I1 rN 1- 0 0 :1 1- C, tLOC 0. , NN I NI ~ )C 00 U', . 'S -1 0, -119- 0 C r~~~~~~~~~~~~~~~~~~~~~~~:; C C0 CD O o c C, C C 0 C CC 00000 o 0 C C 00 0 0'00C- C, OC~s n 0 oo C C0 00000 0 0 0 C,00 C, C 0 0000 C OOOo Oo Co r> OCc C o cDo OG o a cC C 0 0 C' 00 000~) C0,0 000 C 0 OC0 O OC 00 C- o C OC C o o C 00 O) OO tr. rt~~~~~~~~~~~L -1 C 7, 0 -- O ~ ~~ c~~ c ~ ~ ~~~~~~c ;~_~~~~F L-' ~'u--~ " .,..m" . ~~~~~~ C C4 c c ~ C~ ~ ~CCC,r CCC._ CCC. -- r' I P, I Em IC u' . - LL c '--- , : .'C C = C.C WC~ C.' n L' C' I- C' C~~ C r -L C C c~~- - CCC _C C_t- 2 .LC ,r' j C 1 -4 ~. C C C_ LW LC r ^ W11.9< _ ~~~~~~~~~~~~~~~~~~~~~~ -120PROGRAM 2 EXEC FILE 0 0 0 0 c c 000 000006'3,000000 C>C0) '"4D o °3 0000000o c 0 0000C2000000 ' 0 O C cc C,0 O C ,"'"' o C,0 O- 00 ,- 0 o 0a o, 00 0 0 C C v., C;~~~~~~~~ , , , ¢ -: F- ~~~~~~~~~~~:,. D.'- t-,-r v~~ ~ ~ ~ ~ ~. 2 o~~~~~~~~F4 :> > 00 a ,U r- :: C C - .- CG 0 0 cr 0L^ 0 0 C ^* > _ 3 _ : . ,- >o C . ; _' C. 0_ t) =~ acF~_ .- - --- r- D - r--, 2 ..c 4- c N - -3t OOL. v _-- . _ - I -121- 0 0 C0000 : 0 0 0C 0 C C OO C- C c° O O 000000000000000000000000000000000000000 O O CD O) 0 O O a a O N 00000 t c 1r a Co OC0 N C O O) n C O O O= a a C 0 Oa 0 0 CD C n O!' O O ~O O-, O> ,O - N r 000 C '- C,- O C 0 O O, CD i3 ( C OR C, C N ,OO C:> O o O C) 'D C C, O O c 0 0 co 0 00 000 r, °l r,OO 000000 ,0 C O Ò 000 CI ,C CD CO E C 0000 000000 0 OCO C,0 -C 0 o oooooooooooooOoOOooooooOOoooooooCoo CO0 C C'J C -- CD O 0 0000 0o o 0 0 00 0 o 0 o 0 0 0 C 0 C0 0O C00 000 rr *_ C - U u- 0-I' -< LOH r_ -- Lr ~- 0 Y* 0, r,^ 0 2 C H r C' - S F C0 U H H -H L .~~~~~ CC,0 _~~~~~ - y~. U :4 L -000>L H (. _ _ U- U- OU-I_ 1- en~t~U-~00C O . CHH 00- C = C ~ r-_I . 0 - C C F-E- 0- 0 = - - H C,r H =C 0z~ C - Uz- -CI tU C< O CH X H = . z E--~ -0HN L .~G H H0 c '= 3 C, r E-CC r ii * 0 C- .C - C .,r.W N_ r- 0 * C orr- C- C C, ~U ~ O~0 Lr -~ H 00 H - C H ar w ~F F : *C F rL - II L)z,~. H- H 11 tO' UX -0 5 rH -e F O:E ii F H__00 0 0 H .H ,. H - C H'0o'H Co'o _O U:= HGF X N 0 0,0 _ H. . -OH __ U- 0 _ ~. F 4 C .L '0 N9 0 _E r, ~_a g C0 0 N 200 Ur I ., 0 0 I C, _ I .I I .I. I " -~~~~~~~ -122- 0 0 ~,oo C0,C 000000 C' C: 0 CsC 0 000 r ' 0 000 C QO00° w z uu, C - : C 52 ~ r, ~,c 'JCY 0_L vU a c zWC CD - r - *K >- U- _ U _~C C .' _ U 0: a . I I I 1 r~ I ~ ~~~ ,.- ''- - - -a- . I -123- ·I 0 0 CD _- A4 C _ O C" Cc)0bc0rA;c C, 0 0 CD C C 0 ooo coc QC t_ -, C _ - :: O 0 C- C C. _ , C00 C' 0 C oooC7, o o. - 4 I C 0C 00 C000 C,0 0 Coc-7>o : o ,ncc >o o o oo o 0 0'*. 0 0 0 Oa0 S 0 0 (Nt 0t N q 0 O0 O 0 o0oo- O0 C14 <1As N C ° N (. > O0 - 0 C0 -c nc-cc _ 4 m^ o o00A o00 o k~s x m 0 C 0 0 Dc 00 oc 3 C -Inq c r-:::ow0,oc octoo ot o o o LO 2 0 I m C CO m~ _ Ar 0 0 ,,A,Înn 0 .Z r = -. 3 0 Oc T3 -t O0 C O O00000 O O ~cc 00000000 ~cc-c c ,',c 0E -. 0 0 C0 C oO C C_ 2 ) O - 0 D~ ,c ,o O O 0C AA :Z, - -4 '._ II I _ CO _., ,,.j f UD ,,- 0 0 0 . ,>, C C,- C _ 00 C 11 00 0 C FF.. E- 0 Z.7 0 C; OFF. _ * F-- 0 - 0 C* - * a.. 0 C0000 -F-- 0 (I: ti: 0 0 = v: u; Z, - C' - C- L' U F.- F F -.. - I- F- F- C - C 0. C 11 > : r C C F=r = F-0V0' LC K - C_ 110 2:c _I *_4NGUCF_ C 99 ff CC--~ N.Z-4. - _ Ii -~ F3 II L't C -I- 0 or - 0- -U CO _ I 11 .- " Ii CO 00F- F2 U' c- F--__ .F F_r- __ *;o) - _ _ = er C11 - = Z, C Z Z, I . 71 c to L^ -I _0 = *II' O - FJ C C C F- 1 :7.- 0 _ *n U; F-C >c .-- F _ _. <~ ._ .- 11 C "fl, c - C F- ~SL ' . E- _ 0 _ F_- . C F-r- C F- - 0C _~- N, Z' - ,F- CC Fr~C C 000 C 0- 0 0 C C _ 0 _ 11 - C r - ccII - Z' - .j - a F- r -F-C c. u 0 '_' L 1r - C :" - ! _l, , ! 0 C r L 1 I -.. -- -124- N c 0 ' ~(c0 C o CD o C C 000 0 000000 0> 0 :)0 O 00 OC0 '0'0o '000 o000o~~C>00 o o C 000000:DC,-. 0- 0 0)~ 0 ,=o c 0 0 0D 0 -0 0 0 eo o 0L C ,.< E- ,~ I _I c ': i i i i i < C~ < I FI I -- I I, LU - * I 0 1- c0 F-!-\ z' X -~ - _-~ =- -E- .lT F- F- bEC I ) C C: E- ;2 Vo . 0 t r' 1 -125PROGRAM 3 EXEC FILE 0 0 G 0 C o C. O _u C - 00 o; r c ~ C Cr C 0o CO o C t,~ DooCsm Ur oo:0 o -- 00 0 C 0 000 C. 0000.~~~a o o 0 o a0 0 D 0 oc C O C0C 0 C - - 0 0 o 00 oo o_ - O o C 0 o no 2. z 0a00 o o Er, >^~~~~~~o C U:~~~~~~~~~QC r- r ;0 : r r r- Cf 000 r r C0 r. * 0. 0~ UK 0. Xj, r ~C rc 0ir Sr- 0 C0 C C. -C r- r 'N - /n - - 0 r * ; - - c 0. 0. 0 r: n -r > Iv-~. O F- .< rJ U: F - cLa F0 . 9 - F- C *n Lv 1 >C . 0 .s r ; b- F4 4- r4 -~ ~ a :-c~ v r a -r - FLa 0rr r . r1mm-ol '.4 .4" - 7 I - I . 1 -126PROGRAM 3 0 o 0 O r ooc~~ooo 0 CCo, O mC, 0 (O O Ol < OO C O O : GcO O Ol O O 0 0 - n U') cl O C 0 1." U 011 0 0 aO c C 0 0 0 0 0 0a 0 0o O C O O O CO Ol O 00 O> O C O0 O -D Ol O 0 0 0 0 N 0000 I -- uQ0fl(N N N O O= OOCOOO 0 rN CS4 0 0 0 C cl 0 0 0 O 0 00 0 O_(a :T 'n Q_ r- 'Nm ,(N '1 N 0 0 M! C' oOO C 0 00 000 OO 0'000 00 0 o 0 00 00 00 OQOOOOOOOOOOOOOOOOOt 0000000OC 0- 0 0000 0 0 C 0 O 0O 0 0 el C _ co M O O O O Ol OOOOCC°OO°OOOO°OOOCOO°OOO°OOOOOOOr 7 o C, O O0( 0 0 0000 ,C) 0 0 0. 0,0 0 0 0' o 0 C O 0D(0 c OOl4.Cl--T L 4)t-O_-.O 0 0 0 C' o 000 C= C 00000 CO 0 C 0o O 0 0 0O n r~ C C c, 0 00 C> -r, u" r- cl C O 0 0 00 0 0 1 u; O 00 000 C 0o o 0 O -1 - o 000O 00 00000 0 0000 C 0 0 CD 0 _>¢_ , :Lf, 3:_-. y rL Irn T e< _7 0~ C' C C r Cc 0 0 Cc C ODC O 0 0 e 0C0 0 0 C 0 00 00 ZH D7 C -0. 0 - ,rxr Cl ' ~~ X· c L' H C 0 0 0 0 CC 0. ~ L3 ... 0 r~4_ u - * H H -- '.,' F:_ _,-_I_:.. -(7 - ;_ u'- u~ ~. - ~7, >- - '... U.C C' ~ ~~~~~~~C ~ Lr C. _ ~f'~~ ~~~~ -C 0 0 ~ 0 C '" C _ r0 c C I - H H C. U) C.. C ~~ a ;'<~c F- L2 - ( _C --- U (N C H m .r F-l - "0 -: _ _ 0. N I'>C .~~t *r.- C U) -WrX U. 0> 0 o - E- E- H !=;4 0 0 >,,r * C _I C- Cr C- - -N 1 HUS _C F F Et + 1~ Z r-0_+ 0 -C 0 * -C E~ e- - _ C C: P 11 _ O U (; L 1'* -0 I c 0 0 H H 0 oCoC H C I 4 0 o I- 1 C.. I:::a C- C-C e CC C"r 0 H 1'- i : I- F Z II L .rfUU ,.L : '. HH ^:-a C u 0 O LU ii Z - '- CFH 0 00 - H E- . 0 H _._ ~: C -C * E- -u II -0 -IL 0 D2 ~ - - * 0lCF-E Oil C; U cZ-L,_ - - C.: _~~~~~ C - 0 _li H = : II *0 U ~~~~~~~~~~~~~~~~~~~,,-,. tE-- s.(. : < H II HO - 1; W - cr U) U) _ 0 Crs C. C' 0 0 C: U. : ' 0 ; 03 -) )3 ' .) ) A' IA, I -127- N 0 0~ C C C, C0 C Co 0o0 °- C C C C, 1= C, C) C 0 C COo 0 o O- In 0 0000 0 0 0 o C, - C Z, C' 0 0 OoOOCO C O7 o 0 0 C, 0 O 0 C, 0 I 0 C ~ ~~ ~~~C C ID C O- C c 0 0 - 0 0 0 0 0 In Z= In C, 0 0 0 0 00 C 0 C, 0 0 0 C~ C c= C 0 0 0 0 0000000000000 C o C o0 ID OIOC O0 0000 0 C 00 C 000 0 0000000000C000000000000000000000000000000000 0 0 0 0 0 0 0 0 00 0 00 0 ,00 0 0 0 (0 . 0O0 000 O, w x 0 0 C) 0 O- Oc )0 0 0 00 0 0 0 0 Or o 0 0 D 0 O 0 0 O- 0, O l co OOOOo oCCo 0000 , 0 0 0 O C 0 O Cr 0 0 O 0 C a, 0 0' 0 0 C: 0 0 ' cOO o 0 C 0 C, 0 C'OD o > nD C 0 0 0 000000 - ° O C 0 C 0 C, C C 0 0 O = O c DoO n C C C0 C 0- cc OO O 000000 00000 0 C it 0 C H A 11 C" C,, - c: (. - .I _ . E : -. C-- ,-. C C _ . C r- 11 a:"" C_I : C: C: C , C C C C C C C-,. C- C 2. C C:. C 0 FC C C-. C 2. II Cl - C- C [7 2. [.. = (2 IJ F C CC YC : it , Z * -= E 4: hl 77 >- F-r, C .C C C: r . C_ C- V. E- F : IIU - CC - CL 1tt I - C - s t 1~ -C-i C _L C -~C _-C C 0 CC: C _ Z ; Q 2. I U I CEr '- C, C-_ - _-C: - \ - C r-' = C CC ~C11 -u C- C C C 0T C, C- .'v' C ,--C L C Z CC C C C:0 L C - _ C C C- IN C '- C F-c C v;t) 0 CC ' C C I- C C- C: C C 1-4 CC: CC C tr Ca- ,C o C C CC: * 'C C C 0 C: C: .0 0C-'C .INC cC o C C L-, C C- '-C:C:0 ICCF L: C C- F - C ~ C3 - U CC: C !' . CCC C-C-H -'C C C C. C C C C C C: 2. C .C rC _C C * * C C, CC. - '. C_ F tn CX C - C _ C ,~ ~ 0 C ~I 0C - L -T C c=b_! U w C 0 L0 ), U I'd , I ,Z. _ __ O4, i . -128- I) C C) C0og -aC C) C' o C> C E o _c ^ _ EC ~ ~ ~ : 0 O C_.Ch , , 0C) ..C 0 O 0 0 C o 0~ 0 0 0~ 0 0 0 a O C --0 O O '00 O 0O O 00 C u"l C,"~ O.00" C) C C 0 C oO C) C 0 0 0o 0 a 0 0 0 n 000000000000C000CDC00C0C0 SNxxxr-1Emxmxa _~~~~~~~~ s10 O CC ' CC', " 0 OlC C -,4f~.' O CD C, C> Co CD o~C CD o o o=o C~c~CC'C00CC'CO00C)0C'C)0000 O C O U0',, ' - 00000"'000 0 C) 00 CO O O O C O c' 0 O C 0 0 0 0 o U^ D m LN C. r CD ao 0 C1 0rIL, ae :Q 0 00,f C C) O 0 ~ 0 C, " - -" 0 0C)"-'0Clt oCoCoo.ooooooooooooooooooooooooo00C00C)0000CCOC()CC' ' C o 0 0 ,0CC,' C 0 c, Ut ,- C C,: C) O OC ,,.~ ', _t 0 C C C C C,'C , C F2: C F- C F- C. C ._ C 2: A C -)'-'I' 0r... · ,E- FC 2: Ln ° _~ *" 2: o U E- C C r-2Y ::r 2 c c: C ,. C'. F-.~ *U C C _C: * C F: C~ E- X F-C -,-- _ 9-· 2 FF:, C S E. : r- -I . 9 U, _ = :~ F- _ _ I H C I= C L c - -. tt . C F- *. C ;1: - E- : X 5K<2. ..- ~ ~ C' F-' r/2:F-E.CW2 F- - - C E C - .:- f-N C .' F C" C .r tU C..(,... rI.I.Z C ,C -_ Z: r. C _ F-.c F 2: C C _ CF Ct""' C F.~ - i-r: CCCC - :0CC - : r-FCC-C' C C* U Cn C Cc, C CC'CC 11 r. r- I-Ctfh C. 3_ m.~~~~~~3 " i. ' ;XCC-C - C- *F .. - ' L* -- C) * ^'' , 2:2: E--= =Z -C :- C C 2 2: F- C_ U F- =C L- C C C C o r L; C Xr C' c- c C ~- _ CU C U U I) F C: I , - _ _ !- _ CS C t- = ~.'C~F-:.¢,,- C (ow C C-C 11a _ V 2: 2; _: r'-FF-40 ICC * C L C ~ CF an2:.-w C :2 iiru -F-U C C' F- 3: Y tC C IFI C >.L C C 2: r_ F ) _ SCC 'C .~ - * U . L) L. L) t. ¢'>,4 u' I * :.~~~C~r--i· '-i MFM · ·-- --M - -12 9- AC o CD r cc t 0_ F- 'C - C i. U U C::. - QA b0f0 3 I I ·--cI- L·---1a: I nr~~~~~~~~~~~~~~~~~~ _C~~~~~~~~~~~~. _li& Fr I o= C= C C oo 0 - _. -130- 0r ._'- - ::.- , - 0- ~:O a t-a a' 0 o- N PROGRAM 4 o o0 N_ u41~' _ooo0o a' aO C z O 1 -_ . 0 0000D l0 UO rn m I -4-4 'l LU r' \0 X tn cD N 0 4D CO O O a) r, CD -4 (r) co - A cI un, C II I¢ 11 ' Z-0 _J LU C N N -4 o~n r4- -_ (Y 0 0 C') '-0 4 '.44 - - r-- f- .,"F O Ir I-- -, D r- 0:r N 0 o X)0 : C C a' 0% Z C '0C !1 U: ,' U? 4-oo -z - It C ., 0 I ,~ II :1. , ~'- * 0C o Oh. U.P ' - .-0D 0 0 ! n 7L 9-' J9- o J a] --- I Ii Zn' LU 6 ~~~-. NN 0u 0 0:* , Ct- v i~ 0o3 ^31 ~ W II LU LI. 0 01 Iz o a', a'-c o00 a': -zi % z ,i -. 1. I-, 0- _ Z' (:X \ ,\ \ NNNN . <- - X 0t: - 0 -' -- > a' O-0 11 * LN t N > 4 o II~ O.' ~ ~1, _ ¢;-011 0ZvC oa- 00 L0 n' 0 _ - 0 0 I-. - Z 0 Z_ 6.-:6,L.. 4-- O >- LU Q- r I-t-Z--CI _.J (. Oo~-'z: -.. i..c[ Ul) I4% |' '4 'z LUZ aco II On C. ,I ~,,.. 0 1-4 o f w! e. Z * -) L, I- -, -4 IINZ I z XcJ Lr) * I N' 2 . totz :.I,ilO 3' $ m C a'j a' a'i a' O OO NC U _: *_ _4 _7 r %( N _0-z N p- --- -4-J :5 Q-._ N II- - %0, - CCZ" .' (', ON Un- 6' O". - a 0D 0 N C) O 0 _ - -' a -N IIO oa _O-' r _, 9 ¢it C - - O' Ln L', a1o' N_ \J q 4 00 _ - :ZO VCD LJ0I-X -- (X C 0c -J N e, -iI '"%0 ~' -- C) 0 a' O', a' al LL! '-iL tItl- :5* Z oII D o Ul co, C,. Lr 0%D coi r_'n a' 0 a'o0 U)a' a' 0 -0 i0- Z 0r -t xa rc'r -J , a - 1.4 -9- oL C) Dt~ ()' ...-o 1 _o ao ,-.o UlS ' (7ak(7 71 4% ..n- II P II un - IC _e OJ LUi -4 -44 \0 \t Co al N a' C', 1 N c' o- N C - -,, INN ,0_ - 0 *~CD o r%-4 r-' (x - -J I- _ _- I. P Ln CC 0 0 a0 ¢O z 0t-O \ C Mao O',%0 II Lr - - CD -0 a- 0- _ _ I. 4 N a Z . 71 -13 1PROGRAM 5 0 C) W k) C U VI C E- _ l- r. D C C -- C· C N rt F L ·i xN D u C· e-' N 0· - C= C c C- C 2c - ,, C Cc" - r'r~ C C ' . '- C CD ~' N, C C _C _ r _- _ _ C'CCC'o'Cr rf C C C C CC Y C .S . _ CC C'- C C C C _ - N C C -C cr .o a r U:x rCCLCCCN..C)C)C,C iCC'C r. C ~---s C)o*- IC U. C' c iL : C' GC U " ', . C CC_) C) _'- C) C _.Co CCC C "I _CG G-2fG C r, C)( : C iC C - r)C C C' C' C …__t _- C C) CV 7, CcC ~ 1 C II _; C ._Ct' :,Z .e > C F9uCEl E- C r- F-,- CC PE = r" Xc . 'It. = Z" = C1, t t +c -c zr Z . F C , 'A C C C C C , C C QC c.- C ;, C c, 0 3EG .4 LI CCC~~. ~~ C >c: vc=C c11< 11 11 C If t 1.1 C y: I_ CCCC-CC C~~~CCC.r C) C) e)c.U C; U U't -CCCcC^ C _. C C) 'I t F a, .CCC _ c) CCC :) '_ C) J' .N,~ C C ,-CCCNr CC~I L vC C ; tcC.C C C.C'tC C *' C) Us C)) U' C d ,,' C U ¢) C ' . -) C C) U C CC CC - C U' C C F- C C N - = r. C C - F F- P= C C C - CC C N C r~ o= , C-m C- , CG,0: c>x " C C' -) i tr tr CL u' V. n C c- C o) U' C C C U' C ~ =U' C Ce E E- F C rCCC Z.- C C'i C N C Lr C- C ¢)c" m G C C if Cc r G)LI C ,,: C C 'N I" c_ C N b- C "C C _ -c C C c- - ~EClCC GG; .-c rN C c-NN<; ru Ue U - 0 IZX C, _r tr tr U) C: c . oCO C CI I C C- N N C ¢) c -r C C C ' C C) C C C C C -~o-c C C - e c' cc__ C) z c U LL rs . _ C C C Cc Z ,_ I C 9; ~~ ~ ~ ~ -CCL _1 -,r4 F > 1,* +c *r --. X- OC - - _C C C' C NC C "-- …C___c__c r r r C r r ' C r ~ C - .CCC' sr;r ,C ,, r r C C C 4 I A, -13 2- Appendix III REGRESSION OUTPUT 0r N 04 -] 330O '1I minus 5 days delay LI) N 0 N en rs * r MY e -- 0o N u- ut 0,~.= C- Z U) 0 0 -: ^ U 0 I c Ih F- LJ IN CY, ~ ~11N ~ t re r W;U) w U W .C L) ¢ CL Lr ol Lr' t'~ -NN~ , 00 O~ NS I r o0 ' ,e - 0 : n-**r gr1 _ o C,f o It) - .. 0 W- tl 0 : * U) O b5 C_ Ul 1 r CI C I W a r- -C rCN U Z * ~' C 0 Lt 0t V un -* * * C La. C2 4* = * . L = _ .,_ I -: I,,--- 'D C~~~ e I IFe- K F - h >.- LC. .Lr ii et Cr C r 00C~Ca U) rC LO~ ' C IL ,C Lto - rCZ Cr r- :f0= . C - C. H h. -. * Ca* ~ 9 F- o '"C _0 0 N I C 11~CL - U)( C L' <N O '0 -C Q r-=C-0) r - aS -C S.: - C;0 U .) J I ! -13 4I4~~~~ days five ~minus 'G i U I i E,-4 u. ~~~~ ~ ~ ~ ~ ~~~~C '- I -' I~ I _", c 'o N .. ,4 E- o" n i 0 o f ! 40 t~~~~~~~~ waj 'A ;~~~~~~C4 : } 4U~ N CQ ~~~~~~~~~~~~~~~~z HtNb ~~~~~~ c~~u~~ 00 r, '0 8-' ~C *q 11 E-, H Ob. _~ *X ~ - N :X1 ,~ _ - 0 8-v Cl~~~~~~~~" I9 N II b? * A N 3?C X ~" o 'A U: , cz _~ c ~~ P~ C _- ii~~~~', *: * o C ' E I 'A ~ 0s S ;z:C GA- I~~~~~~~~~~~ I '^ - ~ 'A -= _-" = -C ^ :- - * 11 I N ' ul X : 11c: 11 8 K . r-~~~~~~~~~; Xz I 2ra- i *1 * * O ~ O * * _. -'u ,"8. = =:" "" * * G _ U FHv >_ z E ''D ~ rN :> ~-a~~~~~~~~~~~~~~r @- < i Cl ~ t r .. ~ ~ ~ ~ ~~~~t -O- ~~~ ~~ 8- -- Z -- ,, - - 4 _ > ~ ~~~ ~ ~ -- .. - * c. W0 tA e_~~ -~'4 '0r~'-L0-_A 'i 4 _ c 8--1 _ (~> u L -- . i X ;*'4I-| S H C .. - ^ . t) # u_ * F - a - tn c t < : s < tn H HH L _ I I i I II i A4 -13 5- minus five days ul U gui qN a I r U E- ¢' :,a Cc" Et e I:n cs -C X- - a, 000 -C = 'n Ul U . C") . 1. Ul I r! r -3 C: it ~~ U1UX I a3 t, l 1 Cr f. , X, ,0 U ee' 0 U~~~ 2.2 II' ~ (N· , o', 2 C ii C, F- Ul ~ * ~I I , . Ch i'4 C:.. = C .C C, 4f * . _ - ; ,_ I, - ;t r >- z- : _ I 4 ° = _ ., . _ I(N tf .r l , ,'{ ,I _ a_ E_"~ r3,, r,I I." : - _ _/'I Oi - C-D -w t- .2 i4 -11' t, I E-, E-3 ¢ _~ * : -C '-1 I (D z I ·3, ~ a3 e -~ Ft= = - I = C, ("4 I.Z rl r CN !~~~~~~~~~~~~~~~~~~~~~~~~~~ I : I I : i I . _~~~~~~~~Q day delay ~. ,~ ^zero , O ,,i F- . e4 l -- 4 " , ,., r ~ ,-n _ I U3 '^ ,I , t' II E-4 ,t *" ~ . , e t U · ~ rW ~-. -4~~~~~~~~~~~~~~~r >t I~ ¢ E-4 o. C', ' ~ ~* ~ ~ 0~ 0~0 __ o· : - ~ ~ ~ ~ ~ ~ _ c -13 7I ,_ ,,~~~~~~~~~~~ , , , ,- i i i I I ~ '~~~~~~~~~~~~~~~~~~~~~~~ ·. '_" t ,_ I, 0'. U ~ ~ ~ .4 ~ . * t I , ,J ~ ~ fl I * -4 :I_-4 _ ,i -- -C . 4 * "-1 F ! i0.0 _s-c . 0 , i ~ V *- 4 ~ ~*~~~~~~~~c ~ - 0 , _ A0 C*;4 0n0L I 41 / o ~ ~ .4 i .0 ~ , 40 _j =: _ o4 'Z .: .- ., U C - Uc J r* ~" *-. . .; ~ :.1t1 ' i '' -. c* -> | c . J~ * . Zn ,_,Z 'VI, n zL4 n ' .4 1 l 4 0$ * ~ 11NI-4 - ~ U- -i'J" -< - '; , t ~~ = -'- * . -JL N I3 * !^ -w .'.I 0_ t~~~4!- 'z~~~~~~ . C-. . . X- Wu .4 ~ ~ ~ ~ ~ ~ ~ , er~ ~~ ~~ ~ ~ * 0~~~~ *k 0 'C t s, I *'~ Ir 4 I- ^^ ~ · a -13 8- zero day delay ''~~~~~~~~~~~~~~~~ ,~' 3 t, ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~II ., ,C4 -4 1) it. F -4-4- I |*- _-~4 -c <- NQ i <~~~~~~~~ I ,: I {;;; 4 . 0U N I.4 * n*r ~~- * . ''4 * * O , N4 S' O ~ ~~~z 4 f ~~ ~ ~ ~ ~ I_ * ~~~_ ~2 **C..~sF*,'I>40 -. 2 ~ ~ t* , .4 .4 -: .i _ - ~ 2: 0 t .4 _ _ * -4 -4 0: .. sS , 4 -( 4 C1. 2: [S ~ig .I . '1 ' I -< C 2: 0 - 't 2 J1 *'J 214 t ! * UJ <4 z I 00 ; _ n s H *r * ;o . J * . C * 1-_ 4 i E ; E-) U EH z 1 '- 3 ; '-4 I v I2 ~~~:. *J^ _o4* L o ~~~~~~~ O . ,l S I z ; <, 40 2:~~~~~~~~~~m jQ2:fl2: 2 .1 4; H _I ,H ~ -4 C ~1~~~~~~~~~~~~~~~~~~~~~~~~0. _ .44 I ; f. A ; o >4 4;0_ _.o : >1.n QJ =4 I; W^X,.j> s;L:J CZX< 44 O ^ 114 * 4 4 4 n - .e 4J A~~~~~~~ 1.1 44 4 - ,-- ) ~'4 _ '.1- I- - I~~~~~~ - '.4 _4 ^42 . _ t 1O 'l, - - - i~ ~ N 0 ..., ** *, : =: J1LlU1tzZt 4I 1'. . N I t; .. .. I ,; ! i I j 'i I I I * ^J i a., 11 i , .1 -13 9- five day delay I U'~ A ~ ~ . ~ u i J -4 E-A . t, -OI . I 4 F. .- , Ii A _ 00 t~a z~~~~~ a, -, .e'D :~ O- la= ~~~ O n~~~~~ AE N 0 '0' ~~_ 0 O1, * '---z " n '.9- '0 .4 hd t ;'N't C '~U." 4 O%0 I I X,ai M I a 4 UD E" (",1 ,,^ CQ 31:~X -' ';O o~ , ~ -I ~ 02~ 4C OA~ 0 8t w ulOt .' 0 a, t X Fz~~~~~~~ia O. trt * ;4 <r ,) - ,A- C ~ ~~ ~~~~* OC 0 t- -a ~ - ~ ~ a c -~~~~~ _ ^ - *C .. '- Ai ,, C~ OC 11 'z- t~ Q z fa, C=C2 > O I a -> ~~ Z a, C <wC ~ **X * *> o£v - C *t t C. * 2~~~~t X *--q EU .C - ,;j* 11U uCt -C:: tl z-C4 _ a:rr. ai a: - to a, ~o EI .Ct s¢v¢' r;-ca~ + C ~= 0 ]=> 5^ *C -~ i A A A _ c ,t i .1 a a, 140 i,< five days delay :I ':D I U I',~~~~~~~~~~~~~, 9 ~~~~" 0.* 00 0 II I 0 t _t .,1: - II O tX Ii .,~ t4c7 .N ~ ~ ~_u ~ O Cl oU - ~~ ~ ~ 1I Z I q~ o ~:, x-~~~~~~~~~~~~~I -,, ', --. 1,..1 a.l, "U cor-, , CO · a,3 O~.N l _,"xl ,I C. ZCD ON 40 "' tJ ** E~~~~~~~ 03 .:" Z > " ~~ !,' - * ,,W_~~~- . - c - E- >,r > n4O-*'~ uP 1<- >< -.. -> ot l-> Cl-.. *- rZO !E ~ F'-*;t, ¢ O0 I I ;; so E4 '* utIOON * * ** C 4-.' *~~~* c .t *-. =r I-t * - CA: .. r~l-cCri Y ;~ rJ A cF *J, )o z ?,.' ~ ~ _ _ I ,_ rxr -, ,,,- _ 0 * 00r-: _-~' _ * :w.* ~ ~~I) - + _: ¢:t_ r- wi .D N lt Z 0e: 11 ' i~l:: ",rI Ct -0ul_ _i-i, cl1 . - ¢ ~=: .4 111' em ':, zC' -c . -- t~l ~. ~~~~~~~~~~~~~~~~~~~~~~~~~~~c -~~~~~ -_ X IN w .-- = 0 * ;s s . C #- , .^ - r - < s: r ,c , v. _ .jr _ *- v ,r, -141five days delay U Qn ~~~~U, VI t-4 E4 V), .4 a, N aul e r- t !e O o 4 x .I ao ~~~~~~~k,,,t"* . In*, r- co cm~ -C~~~~~~~~~~~~~~~~C F- L4 ~ ~ V) ~ U Cl04 0. M ",~J~~~~ = 'i = r . . ~~ ~~~~'~~~ _% N~~~~~~O I-,I U, 0 D a.N.r .r,--Il'.~, '.90. ,4~ E- ~~ ifl~Q ~ ~ ~ CL F Fz~~~ * ~ ~ =4 ~ ~~ ~ - 5fC~, ~ o _ c: z a ~ ~ ~ : 9 v ;: _ ~ ~ t0 Z 1 l CG C1 z~.C a , Zb c: 1 . .4 * ¢ *, c- * z C N : M¢ _ ; . U << v~~~~~~~tl Ei tt x < E: h c EC E Z 0 n ; E- NC Z " UI4 .J cz t_ C11 11. L4 r- 21 XUw , 4u? C It~E-'Cfl fl74 tA-C * ZOr 4 . , - e .d -4 e-C i~~~~~l eD Fz o I-4 o O 4 ~ . an E ·. . C_.o. - ~b - U-J 11 1 11 N~~~~ C N.4 -- t~ - X ~ L U 6"~~~ .c >. N I It~ ,<-.0 u N.~~i a '~04 Vo ; F 11 > < Fi . e 2 < o I. "~~~~~~~~~~~~. '~~~~~~~2Cd . - -1 4 2- ten days delay j E=4 - N o4 r E- ul. · CD IM = F a 1 rS :I r4F. O O '-O E. Q = .4 113 E- " 3r. 1-4 -4 D C- ow x~~~~,. a. E- C~ I 4 U O- un " W rn : C w 0 z aw C- C l ---Cr "'~. r, C - C , 2c E-. s r ·C, - r.-w., *"<D ,- ~. r , : l tC U. 2 , I I 1r' l,2 - "a _~, _l _ ._ .r r_. - . -C n U" - E-A . *~-~ .. ,_ ' -_ C A1 .. I .. r. -143- -i! ten days delay f x: 0F C ,(, j4 e E4 VI I , . (N mtt 1-4 :1, I 0 Q 3 = Fc i' aY I= = H00 I U, E- t:_ UI e , m . 17 1'.4 - p 0. 0 0 I i4IW C4 E- (-4 I 0zo q it at C) _ 1 V a o 0 00 c ~-4 tX 1 " C2. r(Na a.". (N _it -C - 0i UO-rh i4 tJ' E-4 t- (N a 11 0 Ur ;.- w 0 .12 =- -' =;-E- cm~ '-, Tr' U.*4 t U i, n.U,-t -C IIw 11 z ,, C '-C ,:1, f_ t0 0r ( tr_ = N I I >44 4: ~,, E0 a: I -J Ct 0_ L~" O a. : : . :IC 4_5- r[C z= in m W a.otE a: . _ C UI:~. . I I I. ( Z r-4 l= 4 . 4 (; j IC c c 1 :2X I .4 Q E-, r,, r. :r zr. eE _ _ -. ., . I fl, C7 I.., I..., -. 11 I.., I.., -w -. i Ii I -144- ten days delay U l "J r" 0 o O o: x (N N, O tN In -4 E- .4 0 ; I Q° C 00 Uz ' -C &4w I mr 0 I- ZO0 L' -i z . CY a~ . I=r U 0 n U Ca U a, 5N , CN LM o o E- ~~~~~ U~~~~~~.., (",iN *, U 'M h I II I o _ U, m 0 Un r- m ~ ~ I~~~~~~~L 0NO 0 U E 11 *7 wC U C3 m 0UC E- "i¢. 'C N -- L4 U. .- :r U in f- C 0q '-4 ZXC <C r'- I U, - - . n: * ,, S C -:H9 Fr b i[1- ( - if G 1 ;= C7 ; ¢ I _. (N - C t- .. ¢ w-, :I_ , - Ov P- E- 10 0 t 20 . E. I r" ., U= z I m (" 0 r E- _4 t._ t~ta * ~ ~ cr< - wC < ~ ~ C.~~~~ _. = - t~t C -4 u¢ X : C .. ¢ Ew' - E - C I rv N ~-145- -c<~~~ twenty days delay (J ~~~~~~~~E -U~ 0 " I-. r. 641-~~~~~~~~~~~~~. (% 00~~~~~~~~~~~~~~~ o- a iU ~~ ~-. ~ ~ aM HQ~ cr x :S ~ ~ ~ ~ _ * -2= * -. c a * * -* r_ r ^.Cu)r'Cr< c: r *U1 ~ -4 ¢U 1§ - 4 ~ :-C:< i 44 U) z 0 I/ r:~ c:. -'- C~: ClI z 4 . a.2 * 00 ;* { r < * rI-. .4 H§ z - ¢ U * C U. 0 e 4 -sC ~~~ 04 U> 1- S: O0 40 -< ~ ¢u<: r _ dE- _ 11 > ~xrIE E- ~ _ S Q LJ~C VI ,.>, r ^P ~F*c;ou -~ =k a ' C ' U : ~~~u ,Q ru j, .4 zS _ 0 I. ~~~~ ~~~~~~F * 4: .4 z U1N' ~~~ ~ b ItO m 0C O &n * -A 0 ZN .. CiJ ~ tA ~ 0 -1 Fi =~~~~ cC , * U, *1, _ rr a~~~~ o O ¢ - -_* *,.n : ~I tn* a U 2- ; ~ '-C ~.. , 8 QD01 r2 ~~ rR , 2I 0 I c J - 9, . . * Z: U C I * *m~S * *> @ k -c - > ¢u7 ) C -146- twnety days delay U 0E- l -. , r P. CI 1. E- 05q to In- H .< :n~ x 00~~~ < tD t= cx C ""* r*I - I D ~~ ~ U I ' I U 9, C C4~~~~~~~~~~~~~~~~~~C IU c I z t ~ r ,.N H t. - *_ NI b < [r t - 0- t~~~ H Su && * NO t- -. o- , 11 1 0' C cc C, c} ftl4 c tN * -O * c-e , -t-t r- F' C CCC ~ ~~ ~ s- H C >~~~~ 2, ~ ~ ~ ~ a~~~ : -. ~~~-~ =c ~ c. : I - ' * '~ U . = - -_ = ~ _ tH [ F t S z fq IICIIh. ; r ~-~' u> _ 1LZF9 C CI CCr N *, :e > '* *> I U Z C *c 9 ', U U - :! tc t t~~~~~~~~~~~~~~~;: c = 00 * SZ*OO r<. ' CC C * G- G LF * - - VI t t- C C E , C 9 L C~- > z t ~ C.h C cc : C LoC-U a- < X t -~ - r LO ~~~~~~~~~~- t- t C * * - IIr. , .D ^H - C G ' 11 t i,W ) U) - *- tF F . =L tD - C: C lt _ t _* C..~~~~ _ C_ ,*.,, ~ ~~U _ _I I L C~ - e IF XF ,.F9 - C. ,>,? N L. ^ T I I I -J I .I I II -14 7- ., twenty days delay 0 U Ei- I 0( o Or N r- . 3-4 (1 '" I I. N 0 II, . 01 *^ n u F". 2 Uen H U '9i 0 0 I X rN _ U , _ ,r EH EW W H. ia. U w 00 o o_ - _ I, e. r t IiD0 -, ' : C) rs ~ II U~,, ,,e C~~~~~, 2U . . zL' ..,-Z _., e :4 C = 7 ,e CU = IC Nz N r~ 1 0.1 L U * H Z _*- C00Ce L E; rXtrt 01 5:_ C Ln I . I Cz~- F O ~~~~... . 1-i ill'fi: rL.K I U * n ot <: XX :4 0*' If -z frO = x i I V oo b - 0i i -4: r U¾ i I (-I I - c o 0 0II- ~ C o NœC3, .It o I O V E- . N, i- in * ,, C o ~II I f 0 M ".-C' ; Ft - . rC, 2 C M z ;1_A Im C ;-:. Az, j "p :- rlC Ny -148- GLOSSARY OF TERMS AND EQUASIONS Alpha (). The alpha is the intercept of the regression line on the verticle axis. A positive alpha indicates Value Line has earned, on the average, a premium above that expected for the level of market variability. It's expected value is zero. Beta Beta is the regression coefficient of (). the rate of return on the market in the market molded equasion R1 = + BiRm + i Correlation coefficient (p). This is a measure of the degree to which two variables move together P,2 1- S1.2 .2 = Si2 Covariance (cov,y). Another measure of degree two variables move together. A positive covariance measures on merge the variables move together Z (xi - x) (x - x2 ) N Efficient market. In an efficient market, current security prices fully reflect all available information. Efficient portfolio. A fully diversified portfolio. -149- Geometric mean. A geometric mean is the Nt h root of N observations. Least square regression line. regression A least square line minimizes the sum of the square of the verticle deviations from observations points. xi = a + bxy Risk free rate. The risk free rate is the rate of return on virtually riskless assets, usually Treasury Bills. t-statistic. A measure of statistical significance. An absolute value of 2 or greater is good. -150- FOOTNOTES Chapter 1 Introduction Introduction Samuelson, Paul A. 1. Chapter (35) 1 1. Bernhard (6), pp. 13 2. Bergstrom 3. Bernhard 4. Markowitz 5. Lorie 6. Sharpe 7. Bernhard 8. Letter from Value Line to Fischer Black 9. Sharpe 10. Value Line "Selections & Opinions" November 17, 1978 11. Bernhard (5) 12. Bernhard (5) 13. Malkiel (30), pp. 14. Malkiel (30) (4) (5) (29) Hamilton (29) (37) (5) (37) 107 -151FOOTNOTES Chapter 1. Shelton (42) 2. Shelton (42), pp. 285 3. Shelton (42), pp. 260 4. Hausman (20) 5. Murphy (31) 6. Murphy (31) 7. Kaplan & Weil (26) 8. Kaplan & Weil (26) 9. Black (9) 10. Black (9) 11. Black (9) 12. Black (9) 13. Black (9) 2 -152FOOTNOTES Chapter (16), 3 34 pp. 1. Fama 2. Lorie 3. Lorie and Hamilton (29), pp. 77 4. Sharpe 5. Lorie and Hamilton (29), pp. 89 6. Lorie and Hamilton (29), pp. 89 7. Sharpe (38), pp. 138 8. Jensen (25), pp. 389 9. Lorie 10. Lorie and Hamilton (29), pp. 91 11. Jensen (25), pp. 413 12. Sharpe (41), pp. 29 13. Sharpe (41), pp. 29 14. Sharpe (40), pp. 30 15. Sharpe (40), pp. w7 16. Jaffee (24), pp. 427 17. Jaffee (24), pp. 424 18. Ibbotson 19. Black, (29), pp. and Hamilton 75 (39), pp. 119 (29), pp. and Hamilton (22) Scholes (11) 90 -153- FOOTNOTES .Chapter 6 1. Wall Street Journal, May 8, 1978 2. Your Income Tax, Government Publication 3. Implications of the Capital Gains Tax for Investment Decisions, Hold and Shelton 4. Black 5. Business Week, November 13, 1978, pp. 172 6. Boston Globe, December 8, 1978 (9) -154FOOTNOTES Chapter 1. Shelton (42) 2. Hausman (20) 3. Murphy 4. Kaplan & Well (26) 5. Black (9) 6. Fama (16) 7. Sharpe (39) 9. Jensen (25) (31) 7 -155BIBLIOGRAPHY 1. Akermann, Charles and Keller Werner, "Relative Strength Does Persist!," The Journal of Portfolio Management, pp. 38, Fall 1977. 2. Ambachtsheer, Keith P., "Where Are the Customer's Alphas?," Journal of Portfolio Management, pp. 52, Fall 1977. 3. Baker Securities, "Funds Evaluation Service," Annual Report Text, 11/74 - 10/75. 4. Bergstrom, Gary L., "A Performance Analysis of Pension and Profit-Sharing Portfolios: 1966-1975," presented at the May 1977 CRSP Seminar. 5. Bernhard, Arnold, "Investing in Common Stocks," Arnold Bernhard. & Co., 1975. 6. Bernhard, Arnold, "Portfolio Management: Active or Passive," Presentation for Eighteenth Annual Management Conference of the Graduate School of Business and Executive Program Club, The University of Chicago, March 1970. 7. Bernstein, Peter L., "Efficiency and Opportunity," The Journal of Portfolio Management, pp. 4, Fall 1977. 8. Bishop, E.L.,III, and J.R..Rollins, "Lowry's Report: Denial of Market Efficiency," The Journal of Portfolio Management, pp. 21, Fall A 1977. 9. Black, Fischer, "Yes Virginia, There Is Hope: Tests of the Value Line Ranking System," Letters to the Editor, Financial Analysts Journal, September/October 1973. 10. Black, Fischer and Myron Scholes, "The Effects of Dividend Yield and Dividend Policy on Common Stock Prices and Returns," Journal of Financial Economics, pp. 1-22, 1 1974, North Holland Publishing Company. 11. Black, Fischer and Myron Scholes, "Session Topic: Individual Investors and Mutual Funds ... From Theory to a New Financial Product," Journal of Finance, Vol. XXIX, No. 2, May 1974. 12. Boston Globe, December 8, 1977. -156- 13. Business Week, November 13, 1977, pp. 172. 14. Cannistraro, Richard S., "The Predictive Ability of Price/Earnings Ratios for Determining Risk. Adjusted Performance of Common Stocks," Master's Thesis, December 1973, M.I.T. Advisor - Robert C. Merton. 15. Ehrbar, A.F., "The Trouble With Stocks," Fortune Magazine, August 1977. 16. Fama, Eugene F., "Components of Investment Performance," Journal of Finance, Vol. XXVII, No. 3, pp. 551, June 1972. 17. Fields, Richard L., "Alternative Regulatory Approaches: The Automobile Insurance Market," Bachelor of Science Thesis, Department of Economics, M.I.T., July 1977. 18. Fullerton, David John, "A Further Investigation of the Value to the Naive Investor of the Trading Activity of Corporate Insiders," Master of Science Thesis, M.I.T., May 1978. Advisor - Fischer Black. 19. Garbisch, Michael and Alexander Gordon, "Is Standard and Poor's Master List Worthless?," Journal of Portfolio Management, pp. 34, Fall 1977. 20. Hausman, Warren, Note on Value Line Contest , Journal of Business, pp. 317-320, July 12, 1969. 21. Holmes, John Russell, "100 Years of Investment Experience With Common Stocks," Financial Analysts Journal, pp. 38, November-December 1974. 22. Ibbotson, Roger, "Price Performance of Common Stock New Issues," Journal of Finance, May 1975. 23. Ibbotson, Roger and Rex Sinqufield, "Stocks, Bonds, Bills, and Inflation: Year by Year Historical Returns (1926-1974)," Journal of Business, pp. 11-47, January 1976. 24. Jaffee, Jeffrey F., "Special Information and Insider Trading," Journal of Business of the University of Chicago, Vol. 47, No. 3, pp. 409, July 1974. 25. Jensen, Michael, "Problems in the Selection of Security Portfolios, the Performance of Mutual Funds in the Period 1945-1964," Journal of Finance, pp. 386, 1968. -157- 26. Kaplan, Robert and Roman Weil, "Risk and the Value Line Contest," Financial Analysts Journal, pp. 56, JulyAugust 1973. 27. Kim, Andrew, "Is Indexing Really the Answer?," Journal of Portfolio Management, pp. 65, Fall 1977. 28. Korin, Basil P., "Introduction to Statistical Methods," Winthrop Publishers, Cambridge, Massachusetts, 1977. 29. Lorie, James and Mary Hamilton, "The Stock Market," Richard D. Irwin, Inc., Homewood, Illinois, 1973. 30. Malkeil, Burton G., "A Random Walk Down Wall Street," W. W. Norton & Company, New York, New York, 1975. 31. Murphy, John Michael, "The Value Line Contest: Financial Analysts Journal, pp. 94, May-June 1970. 1969," 32. Murphy, John Michael, "Efficient Markets Index Funds, Illusion, and Reality," The Journal of Portfolio Management, pp. 5, Fall 1977. 33. Myers, Stewart C., "Modern Developments in Financial Management," Holt, Rinehart, Winston, Hinsdale, Illinois 1976. 34. Rivin, John, "The Impacts of Unionization on Unskilled Hospital Workers and the Hospital Industry," Thesis, Department of Economics, Harvard University, March 31, 1977. 35. Samuelson, Paul A., forward "Investment Decision Making," James L. Bichsler, ed., M.I.T., pp. 1, March 1972. 36. Schulman, Evan, "Can the Market Forecast Itself?," Journal of Portfolio Management, pp. 57, Fall 1977. 37. Sharpe, William F., "Investments," Prentice-Hall, Inc., Englewood Cliffs, New Jersey 07632. 38. Shannon, Donald, Keith Johnson and Gregory Neal, "How to Beat Those Index Funds," Journal of Portfolio Management, pp. 28, Fall 1977. 39. Sharpe, William F., "Mutual Fund Performance," Journal of Business, #39, pp. 119, 1966. 40. Sharpe, William F., "Likely Gains from Market Timing," Financial Analysts Journal, March-April 1975. -15841. Sharpe, William F., "Adjusting for Risk in Portfolio Performance Measurement," The Journal of Portfolio Management, pp. 29, Winter 1975. 42. Shelton, John P., "The Value Line Contest: A Test of the Predictability of Stock-Price Changes," Journa of Business, 40, pp. 251-269, July 1967. 43. Stearns, Linhart, "A Modest Proposal," Journa Portfolio Management, pp. 70, Fall 1977. of Timbers, Stephen, "The Non-Efficient Market is Not 44. For Institutions," Journal of Portfotio Management, pp. 59, 1977. 45. Treynor, Jack L., "How to Rate Management of Investment Funds," Harvard Business Review, pp. 63, JanuaryFebruary XLIII. 46. Villani, Edward, "The Theoretical Challenge to Index Funds," The Journal Portfolio Management, pp. 46, Fall 1977. -~~~- MITLibraries Document Services Room 14-0551 77 Massachusetts Avenue Cambridge, MA 02139 Ph: 617.253.5668 Fax: 617.253.1690 Email: docs@mit.edu http://libraries.mit.edu/docs DISCLAIMER OF QUALITY Due to the condition of the original material, there are unavoidable flaws in this reproduction. We have made every effort possible to provide you with the best copy available. If you are dissatisfied with this product and find it unusable, please contact Document Services as soon as possible. Thank you.

MITLibraries

Related documents

Products

Support

MITLibraries

Related documents

Add this document to collection(s)

Add this document to saved

Suggest us how to improve StudyLib