3/23/00 252y0022 ECO252 QBA2 SECOND HOUR EXAM March 21,2000 Name ______key__________ Hour of class registered _____ Class attended if different ____ Show your work! Make Diagrams! I. (14 points) Do all the following. x ~ N 10,6 9 10 1 10 z P 1.83 z 0 .4664 1. P1 x 10 P 6 6 11 10 11 10 z P 3.50 z 0.17 2. P11 x 11 P 6 6 P3.50 z 0 P0 z 0.17 .4998 .0675 .5673 0 10 10 10 z P 3.33 z 1.67 3. P10 x 0 P 6 6 P 3.33 z 0 P 1.67 z 0 .4996 .4525 .0471 15 10 4. F 15 (The cumulative probability up to 15) Px 15 P z 6 Pz 0.83 Pz 0 P0 z 0.83 .5 .2967 .7967 17 .5 10 14 10 z 5. P14 x 17 .5 P P0.67 z 1.25 6 6 P0 z 1.25 P0 z 0.67 .3944 .2486 .1458 6. A symmetrical interval about the mean with 44% probability. We want two points x .72 and x , 28 , so that Px.72 x x 28 .4400 . From the diagram, if we replace x by z, P0 z z.28 .2200 . The closest we can come is P0 z 0.58 .2190 . So z .28 0.58 , and x z.28 10 0.586 10 3.48 , 13 .48 10 6.52 10 z or 6.52 to 13.48. To check this note that P6.52 x 13.48 P 6 6 P0.58 z 0.58 2P0 z 0.58 22190 .4380 .44 x.075 We want a point x .075 , so that Px x .075 .075 . From the diagram, if we replace x by z, P0 z z.075 .4250 . The closest we can come is P0 z 1.44 .4251 . So z.075 1.44 , and x z.075 10 1.446 10 8.64 , or 18.64 . 7. 18 .64 10 To check this note that Px 18 .64 P z Pz 1.44 6 Pz 0 P0 z 1.44 .5 .4251 .0741 .075 3/23/00 252y0022 II. (6 points - 2 point penalty for not trying) A manufacturer of mowers wants to compare the amount of time it takes to mount a mower engine using two different processes. Two independent samples are taken and the time in minutes appears below. x1 2.1 4.1 9.1 3.1 2.1 x2 3.2 7.2 5.2 8.2 4.2 3.2 Note that x1 4.1 and s1 2.915476 s 2 1 8.5 a. Compute the sample standard deviation for the second process, s 2 . Show your work.(3) b. Compute a 95% confidence interval for the difference between the population means 1 and 2 assuming that the variances are equal for the two parent populations. According to your confidence interval, is there a significant difference between the population means (You must tell why!)? (3) Solution: a. x 2 5.2 and s 2 2.097618 x s 2 2 4.4 x 22 10.24 51.84 27.04 67.24 17.64 10.24 184.24 x2 3.2 7.2 5.2 8.2 4.2 3.2 31.2 31 .2 x2 5.20 n 6 x22 nx22 184 .24 65.20 2 2 s2 n 1 5 4.40 . s 2 2.097618 . 2 b. From page 10 of the Syllabus Supplement: Interval for Confidence Hypotheses Interval Difference H 0: 0 d t 2 sd between Two H 1: 0 1 1 Means ( sd s p 1 2 n1 n2 unknown, variances DF n1 n2 2 assumed equal) Test Ratio t sˆ 2p sˆ 2p n1 1s12 n2 1s22 n1 n2 2 DF n1 n 2 2 5 6 2 9 n1 1s12 n2 1s22 = 48.50 54.40 34 .00 22 .00 sd s p n1 n2 2 1 1 n1 n2 9 9 6.22222 6.22222 1 1 6.22222 .366667 5 6 d cv 0 t 2 sd d 0 sd x1 4.1, s12 8.50, s1 2.915476 x 2 5.2, s 22 4.40, s 2 2.097618 d x1 x 2 4.1 5.2 1.1 Critical Value .05, 9 t.025 2.262 2.28148 1.51046 2 3/23/00 252y0022 Confidence Interval: d t sd 1.10 2.262 1.51046 1.10 3.417 or -4.51 to 2.31. The 2 interval includes 0, so there is no significant difference between the means. H 0 : 0 H 0 : 1 2 H : 2 0 Formally, our hypotheses are H 1 : 0 or or 0 1 We do not reject H 0 . H 1 : 1 2 H 1 : 1 2 0 1 2 3 3/23/00 252y0022 III. Do at least 3 of the following 4 Problems (at least 10 each) (or do sections adding to at least 30 points Anything extra you do helps, and grades wrap around) . Show your work! State H 0 and H 1 where appropriate. Use a 95% confidence level unless another level is specified. Note: Most of you didn't bother to tell me what your hypotheses were in any problem but the first one. This means that I couldn't really tell whether your critical values etc. were right or wrong! 1. For your convenience, the data from the previous page, giving samples of the time to mount a mower engine, is copied below. Test the hypothesis that the mean time to mount a mower is greater for method two than method one. x1 2.1 4.1 9.1 3.1 2.1 x2 3.2 7.2 5.2 8.2 4.2 3.2 Note that x1 4.1 and s1 2.915476 s 2 1 8.5 a) Choose the correct null hypothesis from the list below and write the alternative hypothesis. (2) H 0 : 2 4.1 H 0 : 1 2 H 0 : x1 x 2 H 0 : 2 4.1 H 0 : 1 2 H 0 : x1 x 2 H 0 : 2 4.1 H 0 : 1 2 H 0 : x1 x 2 H 0 : 2 4.1 H 0 : 1 2 H 0 : x1 x 2 H 0 : 2 4.1 H 0 : 1 2 H 0 : x1 x 2 b) Assume that 1 2 and find a critical value appropriate for this problem, using a confidence level of 99%.(3) c) Use your critical value to test the hypothesis. State clearly whether you reject the null hypothesis. (2) d) Repeat the test using (i) a test ratio (2) and (ii) a confidence interval. (2) find an approximate p-value for your null hypothesis. (1) e) Do a 95% confidence interval for the ratio of the standard deviations of the populations from which x1 and x 2 come. On the basis of this comparison, what assumption should we have made about the equality if the variances? (3) Solution: a) The hypothesis says that 1 2 . Since this does not include an equality, it is the alternate H 0 : 1 2 hypothesis. The null hypothesis is the opposite, 1 2 , so that our hypotheses are or H 1 : 1 2 H 0 : 1 2 0 H 0 : 0 or where 1 2 . Did you make the same mistake on this exam as H 1 : 1 2 0 H 1 : 0 you did in the first one! 4 3/23/00 252y0022 s b) .10 , x1 4.1 and s1 2.915476 We have already computed sˆ2p 2 1 8.5 , x 2 5.2 and s 2 2.097618 n1 1 n2 1 n1 n2 2 s12 DF n1 n2 2 5 6 2 9 sd s p s22 1 1 n1 n2 = s 2 2 4.4 . 48.50 54.40 6.22222 9 6.22222 1 1 1.51046 5 6 9 d x1 x 2 4.1 5.2 1.1 . t.01 2.821 Critical Value: d cv 0 t 2 sd is the formula for a 2-sided critical value, but since our alternative hypothesis is H 1 : 0 or H1 : 1 2 0 , we want a critical value 9 below zero and to use t9 t.01 . So d cv 0 t s d 0 2.8211.51046 4.261 . c) d 1.10 is above this value, so we do not reject H 0 . Unless you gave me a clear indication of what the rejection region was, e.g. a diagram, and unless the reject region did not include d 0, you did not get full credit on c) or d) . d) (i) Test Ratio: t d 0 1.10 0 9 0.728 . Since this is above t.01 2.821 , sd 1.51046 we do not reject H 0 . (ii) Confidence Interval: The 2-sided interval is d t sd , but since our alternative 2 hypothesis is H 1 : 0 , we use d t s d 1.10 2.8211.51046 1.10 4.261 3.161 . Since 3.161 does not contradict H 0 : 0 , we do not reject H 0 . 9 9 0.703 and t .20 0.883 , (iii) t 0.728 . This value of t is between t .25 so we can say that .20 p value .25. e) A 2-sided confidence interval is s12 F DF2 , DF1 s 22 2 12 22 s12 1 s 22 s12 s 22 FDF1 , DF2 F DF1 , DF2 2 22 12 s 22 s12 1 DF2 , DF1 or F 2 , and s12 8.5, s 22 4.4, n1 5 and n 2 6 . So DF1 n1 1 4 2 4,5 5.19 and and DF2 n2 1 5 . .10 , which means that F DF1 , DF2 F.05 2 5, 4 6.26 . If we use the first formula, F DF2 , DF1 F.05 2 2.6866 22 12 2 4.4 5.19 22 4.4 1 reduces to 8.5 1 8.5 6.26 0.08269 and, if we take square roots, 1.639 2 1 0.288 . If we use the second formula, 2 2 8.5 6.26 12 8.5 1 , 12.0932 12 0.37222 and 3.477 2 0.610 . Since either of these 4.4 2 4.4 5.19 2 1 intervals includes 1, we can say that the variances are not significantly different. 5 3/23/00 252y0022 2. The New York Times reported on exit polls taken of voters in 6 Republican primaries. The numbers and per cent that voted for George W. Bush in four of these are shown below. State Massachusetts California Connecticut New York 485 1361 968 1060 Sample Size 262 857 542 604 Number for Bush .540 .630 .560 .570 Proportion a)Test the hypothesis that the proportion for Bush in Connecticut was higher than the proportion in Massachusetts. (4) b) Test the hypothesis that the proportion for Bush was the same in each state. (7) Solution: a) From Table 3 of the Syllabus Supplement: Let p1 be the proportion in MA and p3 be the proportion in CT. Interval for Difference between proportions q 1 p Confidence Interval p p z 2 sp Hypotheses H 0 : p p0 H 1 : p p0 p 0 p 01 p 02 or p 0 0 p p1 p2 s p p1q1 p2 q 2 n1 n2 Test Ratio z Critical Value pcv p0 z 2 p p p 0 p If p0 0 p If p 0 p p0 q 0 1 n1 1 n2 n p n2 p2 p0 1 1 n1 n 2 p01q 01 p02 q 02 n1 n2 Or use s p H 0 : p1 p 3 H 0 : p1 p 3 0 262 542 p1 .540 , p 3 .560 , or 485 968 H : p p H : p p 0 3 3 1 1 1 1 262 542 n3 p 3 n 4 p 4 485 .540 968 .560 .55334 , p p1 p3 .020 , p 0 485 968 n3 n 4 485 968 H 0 : p 0 Same as H 1 : p 0 .05, z 1.645 . p p0q0 1 n1 1 n3 .55334 .44666 1 485 1968 .00076671 .0276572 (Only one of the following methods is needed!) p p0 .020 0 0.7231 This is above -1.645. (Diagram!) Test Ratio: z p .0276572 or Critical Value: pcv p0 z p 0 1.645 .0276572 .0455 p .020 is above this value. or Confidence Interval: p p z s p where sp p3q3 p4q4 .540 .40 .560 .440 .00076671 0.0276895 . So n3 n4 495 968 p .020 1.645 0.0276895 0.0256 . The interval includes 0. In all cases do not reject H 0 . 6 3/23/00 252y0022 b) DF r 1c 1 13 3 H 0 : Homogeneousor p1 p2 p3 p4 H1 : Not homogeneousNot all ps equal O For MA CA 262 857 Against 223 504 Total 485 1361 E For .2053 7.8147 CT NY Total pr 542 604 2265 .5847 426 456 2609 .4153 968 1060 3874 1.0000 MA CA 283 .58 795 .78 Against 201 .42 565 .22 Total 485 .00 1361 .00 CT NY Total pr 565 .99 619 .78 2265 .5847 402 .01 440 .22 2609 .4153 968 .00 1060 .00 3874 1.0000 The proportions in rows, p r , are used with column totals to get the items in E . Note that row and column sums in E are the same as in O . (Note that way is needed.) O2 O E E 262 283.58 242.062 223 201.42 246.892 857 795.78 922.930 504 565.22 449.411 542 565.99 519.027 426 402.01 451.422 604 619.78 588.622 456 440.22 472.346 3874 3874.00 3892.712 2 18.712 is computed two different ways here - only one OE -21.5800 21.5800 61.2200 -61.2200 -23.9900 23.9900 -15.7800 15.7800 0.0000 O E 2 3892 .712 3874 18.712 O2 n E E Since this is more than 7.8147, reject H 0 . (Diagram!) 7 O E 2 E 1.64220 2.31207 4.70970 6.63084 1.01684 1.43161 0.40177 0.56565 18.711 3/23/00 252y0022 3. We have the following data: x 970 , 60 , n 225 and 1 .99 a) If we test the hypothesis that the population mean is 975, what is the p-value for this hypothesis? (2) b) If we use a significance value of 10% and the p-value in a), would we reject the null hypothesis? Why? (1) c) Find the critical values of x necessary to test the hypothesis in a) if the significance level is 10%. (2) d) Do a power curve for the test in c. (6) e) A table of solid-waste generation rates for countries classified by level of development was assembled . We are hypothesizing that as countries' incomes rise they become more alike in their tendency to generate solid waste. For 5 industrial countries the standard deviation for a measure of solid waste generation was 0.0652 and, for 7 middle-income countries, the standard deviation was 0.1728. State an appropriate hypothesis and test it. (4) Solution: a) From Table 3 of the Syllabus Supplement: Interval for Confidence Hypotheses Interval Mean ( x z 2 x H0 : 0 Known) H : 1 Test Ratio z 0 Critical Value x 0 x xcv 0 z 2 x H 0 : 975 x 0 970 975 60 x 4 z 1.25 x 4 n 225 H1 : 975 p value 2Pz 1.25 2.5 .3944 2.1056 .2112 b) If .10 , p value , so do not reject H 0 . c) .10 z 2 z.05 1.645 xcv 0 z x 975 1.6454 978 6.58 or 968.42 to 981.58. 2 d) The table below is the same as used in Problem C2 and summarizes our results and further calculations. xcvL 968 .42 means a lower critical value and xcvU 981 .58 means an upper critical value. These become z L and zU . Note that when 1 0 power , and when 1 x cv Power 50 % . These do not have to be computed. Power 1 zL zU x cvL 1 x cvU 1 1 x 975 968 .42 975 4 968 .42 978 .3 4 x 981 .58 975 4 981 .58 978 .3 4 -1.645 1.645 .4500 + .4500 = .9000 10.00% -2.47 0.82 .4932 + .2939 = .7871 21.29% 981.58 (Same as 968.42) 968 .42 981 .58 981 .58 981 .58 -3.29 4 4 0.00 .4995 50.00% 984.9 (Same as 965.1) 968 .42 984 .9 4 981 .58 984 .9 -4.12 4 -0.83 .5000 .2967 .2033 79.67% 988.2 (Same as 964.8) 968 .42 988 .2 4 981 .58 988 .2 -4.945 4 -1.655 .5000 .4400 .0600 94.00% 978.3 (Same as 971.7) 8 3/23/00 252y0022 Diagram: H 0 : 12 22 e) H 1 : 12 22 s 22 s12 0.1828 2 0.0652 2 7.8606 . DF2 n 2 1 7 1 6 and DF1 n1 1 5 1 4 . 6,4 4.01 , F 6,4 6.16 , F 6,4 9.20 and F 6,4 15 .21 , if you used a significance level of .10 Since F.10 .05 .025 .01 or .05 do not reject H 0 . But if you used a significance level of .025 or .01, reject H 0 . 9 3/23/00 252y0022 4. A hospital administrator wishes to compare the distribution of unoccupied beds in two hospitals. Because she believes that the distributions are quite badly skewed she does not use a method based on means but instead ranks the entire sample and uses a test based on these ranks. Data is below. Hosp 1 rank Hosp 2 rank Difference Beds Beds Day d x1 x2 1 6 3 34 16 -28 2 38 17 28 12 10 3 3 1 42 20 -39 4 17 9 13 6 4 5 11 5 40 19 -29 6 30 13 31 14 -1 7 15 7 9 4 6 8 16 8 32 15 -16 9 25 10 39 18 -14 10 5 2 27 11 -22 Sum 79 131 Note: The two rank sums were not in the original problem. a. On the basis of the rank sums (and assuming that the data represents two independent samples,) test the hypothesis that the median number of beds unoccupied differs for the two hospitals. (5) b. A statistician claims that her method is inappropriate because she has ignored the fact that each row corresponds to a specific day, even if the days were chosen randomly. Rerank the data to account for the fact that it is cross-classified and repeat the analysis. (5) c. Assume instead that this data is normally distributed paired data, test the hypothesis that the mean number of beds unoccupied differs. You may need some of the following data: x1 16 .6, s1 11 .42 , x 2 29 .5, s 2 10 .99 , d 12 .9 and s d 16 .90 . (4) Solution: a) Wilcoxon-Mann-Whitney Method H 0 : 1 2 we get TL 75 and TU 135 . Check: 75 135 210 H 1 : 1 2 . If we sum the two rankings, 20 21 . For a 5% two-tailed test with 2 n1 n 2 10 , Table 6 says that the lower critical value is 79. The lower of the two rank sums, W 75 is less than this value, so reject H 0 . b) Wilcoxon Signed rank test for paired data. H 0 : 1 2 H 1 : 1 2 . difference rank -28 8 10 4 + -39 10 4 2 + -29 9 -1 1 6 3 + -16 6 -14 5 -22 7 If we add items with + and – signs separately, we find T 9, T 46 . To check this, compute T T 9 46 55 10 11 . From Table 7 with 2 smaller T is above the critical value, do not reject H 0 . 10 n 10 , TL TL .025 8 , and since 9, the 2 3/23/00 252y0022 c) Test of equality of means for paired data. H 0 : 0 H : 2 H : 2 0 or 0 1 1 2 or 0 1 H1 : 0 H 1 : 1 2 H 1 : 1 2 0 s 16 .90 9 sd d 5.34425 , DF n 1 9, t.025 2.262 n 10 Test Ratio: t d 0 12 .9 0 2.413 sd 5.34425 d 12.9, sd 16.90, This is not on the interval between –2.262 and +2.262. or Critical Value: d cv 0 t s d 0 2.262 5.34425 12.08 2 d 12 .9 is not on this interval. or Confidence Interval: d t 2 s d 12 .1 12 .08 or –24.08 to -0.02. This interval does not include 0. With all methods reject H 0 . Note that this method is more powerful than the one in c. However, it still should not be used unless the conditions justify it. 11 3/23/00 252y0022 Extra credit. Repeat the following sections of problem III 1) assuming that 1 2 . . Test the hypothesis that the mean time to mount a mower is greater for method two than method one. b) Find a critical value appropriate for this problem, using a confidence level of 99%(4) c) Use your critical value to test the hypothesis. State clearly whether you reject the null hypothesis. (2) d) Repeat the test using (i) a test ratio (2) and (ii) a confidence interval. (2) Solution: b) From the formula table; Interval for Confidence Interval Hypotheses Test Ratio d t sd H 0 : 0 H 1: 0 t Difference between Two Means( unknown, variances assumed unequal) 2 s12 s22 n1 n2 sd DF s12 s22 n 1 n2 1 2 Same as H 0: 1 2 2 2 n1 n1 1 s 22 2 n2 n2 1 .10 , x1 4.1 and s1 2.915476 d cv 0 t 2 sd d 0 sd H 1: 1 2 s12 Critical Value s 2 1 if 0 0 8.5 , x2 5.2 and s 2 2.097618 s 2 1 4.4 . n1 5, H 0 : 1 2 H 0 : 1 2 0 H : 0 or or 0 where 1 2 . n 2 6. . H 1 : 1 2 H 1 : 1 2 0 H 1 : 0 s12 8.5 1.70000 n1 5 s 22 4.4 0.73333 n2 6 sd s12 s 22 2.43333 1.55991 n1 n 2 d x1 x2 4.1 5.2 1.1 s12 s 22 2.43333 n1 n 2 DF s12 s 22 n1 n 2 2 2 2 n1 n2 n1 1 n2 1 s12 s 22 2.43333 2 1.70000 2 0.73333 2 4 7 7.13 , so use 7 degrees of freedom. t.01 2.998 5 Critical Value: d cv 0 t 2 sd is the formula for a 2-sided critical value, but since our alternative hypothesis is H 1 : 0 , we want a critical value below zero. So d cv 0 t s d 0 2.998 1.55991 4.677 c) d 1.10 is above this value, so we do not reject H 0 . d) (i) Test Ratio: t we do not reject H 0 . 12 d 0 1.10 0 7 0.7052 . Since this is above t.01 2.998 , sd 1.55991 3/23/00 252y0022 (ii) Confidence Interval: The 2-sided interval is d t sd , but since our alternative 2 hypothesis is H 1 : 0 , we use d t s d 1.10 2.998 1.55991 1.10 4.6766 3.5766 Since 3.5766 does not contradict H 0 : 0 , we do not reject H 0 . 13