Analytical Statistics Schedule & Syllabus V31.0020 Spring 2010 Instructor: James B. Ramsey Room 806, 19 West 4th. St.. James.ramsey@nyu.edu; 8-8947 Office Hours: By appt. Monday/Wed: 11.00 Room 401, Silver Bldg. Text: The Elements of Statistics: with applications to economics and the social sciences, James B. Ramsey, Duxbury Press, 2002 Warning: You will have been cheated if your text does not contain the c.d, “The Elements of Statistics” that is wrapped on the back cover. Supplementary Reading [books are on reserve in Bobst] : The Skeptical Environmentalist, Bjorn Lomborg, GE 149.L65, 2001 Global Crises, Global Solutions, edt. by Bjorn Lomborg, JZ 1318.G555, 2004 How to Lie with Statistics, Darrell Huff, HA 29 H82 Recitations: Friday 11.00 in room 261 Meyer Bldg. Friday 2.00 in room 500 Silver Bldg. TA: Jonathon Thong, jct292@nyu.edu Room 821, 19 West 4th. St. Office Hours: Tuesday: 10.00-11.00 Thursday: 1.00-2.00, or by appt. Course Instructions Course consists of reading the text, attending lectures, lab. sessions, and homework; each segment complements the others. Because statistics is a subject that has to be practiced to be understood the lab sessions and the homework are a critical component of your learning. Lectures will not necessarily cover every topic in the text, but may cover material not in the text. In class and on average a typical examination question will be reviewed so that you can learn how to answer questions in this class. All examination questions are analytical requiring an essay in response, not mere calculation. You are responsible for all material covered in lectures or in the assigned reading. You are expected to be familiar and comfortable with the mathematics summarized in Appendix A in the text. Exercises and an answer sheet are provided to test your understanding of this material. It is important that you become comfortable with this material immediately. Appendix B contains instructions for learning how to use the labs on your pc or other Windows compatible computer. Instructions to help you get started will be given in the first lab session, but you should have no trouble in using the software, just follow the instructions exactly as stated; remember legions of students have successfully come before you. Information on how to install the software is contained in the handout with the course schedule as well as the list of typographical errors. PC or MAC Instructions Bring your p.c. and the program disc that is wrapped with the text to the first lab. session, January 22. You will learn how to use the software in this first lab. If you are a Mac user, your alternatives are to: 1] If your Mac can process Windows, you will have no problem installing the software; 2] Use the IT lab on the 6th floor of Bobst and either save your output on a memory stick, or use the IT’s Files 2.0 system; 3] borrow a p.c. for the course; 4] install VM Ware, Fusion for translating Windows to the Mac operating system, Fusion is inexpensive, but Windows is not. Invitation: You are encouraged to look over the web site that supports the text with links to data and poses an interesting variety of statistical questions. Further, the site contains a complete list of typographical errors with corrections; see http://www.econ.nyu.edu/user/ramseyj/textbook/main.htm. Advice to the Student: Carefully read the section in the Text headed “Addendum for the Reader”, pages 14 to 17 and be guided by the suggestions. You might also read with benefit the “Preface.” In both cases you will learn what is expected and what is different about the structure of this course. Project There will be a graded project, the topic to be chosen from the list that is attached to this course schedule. The main objective of the project, besides giving students research experience, is to learn how the use of statistical theory and empirical analysis will enable you to evaluate claims that are purportedly buttressed by “facts” and to assess the degree of reliability of cited results. Students in small teams, i.e. not to exceed six, will pick a topic from the attached list. The project is to update and critique the results presented by Lomborg; dispute, refine, and extend where possible his statistical analysis, or pose new, but related questions. Students are to pick a topic and form their teams by February 26 at the latest. Record your topic, the names of the team members, and of the team leader with the TA. Each student is personally responsible for ensuring he/she is in some team. The results are to be written up in a formal paper that is not to exceed 10 pages including bibliography, but not including graphs and tables. Students must pay strict attention to acknowledging in detail their references and data sources. On the last day of class, Tuesday , May 4th. , teams will present their papers to the class using handouts and audio-visual equipment where useful. Presentations cannot last more than 12 minutes. Labs and Mini-quizzes You are expected to attend each lab. session as there will be a short mini-quiz for about five minutes each week. Some weeks proofs and derivations will be reviewed, so be sure to attend. The objective of the mini-quiz is to provide you and your instructors with a quick check on your understanding of the material. If difficulties are observed, they can quickly and easily be rectified and not left until the occurrence of a formal exam, even more reason to attend the labs very conscientiously. Distribution of Grades: Homework completion and lab. attendance: Better of Mid-term I, or II: Project: Final: 20% 30% 10% 40% The final examination and mid-terms will be given as scheduled.; there are no exceptions. If you miss an examination, you will be graded as zero for that examination. For your final grade, the higher of the two mid-term grades will be used. The Final is May/11/09, be sure that you will be there. Topics to be Covered in the Course. The topics to be covered in the text include descriptive statistics with interpretation of the fourth sample moment and sample covariance with special reference to describing the shape of histograms; the development of probability theory as the theory of random variables and probability distributions both discrete and continuous; moment generating functions; an introduction to the principles of inference, both estimation and the testing of hypotheses with the development of the required distributions, Chi-square, Student’s T, F, etc.; the theory of the generation of bivariate distributions is developed including the relationship between joint and conditional distributions; linear regression is explored in some detail along with the analysis of variance. All derivations are supported by Labs designed to illustrate the concepts under discussion. The accompanying c.d. contains two additional chapters that the reader is urged to at least sample; “Nonparametric Measures” and “Bayesian Inference.” Objectives for the Homework Assignments: You should note that the homework assignments in the text are in three groups; “Calculation Practice,” “Exploring the Tools,” “Applications.” Each week you will be asked to complete questions from each group. This is because they have distinctly different objectives. See the comments in the Preface. Notice that many exercises are headed by the statement: “Worked. Objective: Text….”. This is to ensue that you recognize the objective of the exercise and therefore what you are expected to learn; pay particular attention to these exercises, even if not formally assigned. These exercises aid you in fully understanding the intuition and subtleties of the concepts and relationships discussed in the text. Calculation Practice: Efficiently computing statistics is a skill that has to be developed just as is the case for any other skill; you need to practice what should become by the end of the semester routine calculations. You will discover that acquiring this skill will facilitate your understanding, much less, much less your ability to answer examination questions. Exploring the Tools: These questions are designed to enhance and deepen your understanding of the many concepts introduced in the course. Faithfully fulfilling the questions in this section will enable you to proceed beyond routine applications of a concept and will enable you to interpret more precisely the meaning and interpretation of your calculations. Applications: These exercises illustrate the use of the tools you are learning in practical applications. In particular, attempting to answer these questions will guide you to a more nuanced interpretation of the processes that you are acquiring and will improve your understanding of the strengths and weaknesses of the statistical tools that you are learning. Each of these sections is important in itself and is an aid to mastering questions in the other sections. The least you should do each week is to answer questions from all three sections. Do the Homework on time each week! Course Schedule Spring Semester 2009 Date Jan/20 Jan/25 Jan/27 Feb.1 Feb/3 Feb/8 Feb/10 Text Reading Chapt. 1,2 Chapt. 3 Chapt. 4 Chapt. 5 Chapt. 6 Chapt. 6 Chapt. 7 No class; President’s Day Feb/15 Feb17 Feb/22 Feb/24 Chapt. 7 Chapt. 8 Chapt. 8 Mar/1 Mar/3 Mar/8 Mar/10 Mid-term I: material to Chapt. 7 Chapt. 9 Chapt 10 Chapt. 10 Mar/15 to Mar/20 Mar/22 Mar/24 Mar/29 Mar/31 Spring Break: No Classes Chapt. 10 Chapt. 11 Chapt. 11 Chapt. 12 April/5 Mid-term II: material to Chapt. 11 April/7 April/12 Chapt. 13 Chapt. 13 April/14 April/19 April/21 April/26 April/28 May/3 May /4 Chapt. 13 Chapt. 14 Non-parametric Statistics Bayesian Statistics Bayesian Statistics Chapt. 15 Class Presentations Analytical Statistics Schedule & Syllabus V31.0020 Detailed Homework Assignments Due Date* Jan/22 Relevant Chapts. Appendix A & B Jan/29 Chapt. 3, 4 Feb/5 Chapt. 5 Ex. 5.4, 5.7, 5.8, 5. 13, 5.24, 5.32 Feb/12 Chapt. 6 Feb/19 Chapt. 7 Ex. 6.4, 6.10, 6.13, 6.15, Ex. 6.36 Ex. 7.7, 7.13, 7.29, 7.31, 7.55 Feb/26 Chapt. 8 Ex. 8.3, 8.15,, 8.21, Mar/5 Chapt. 9 Ex. 9.3, 9.18 Mar/12 General Review for Chapters 1 to 9 Mar/26 April/2 April/9 Chapt. 10 Chapt. 11 Chapt. 12 April/16 April/23 April/30 Ex. 10.3, 10.5, 10.20, 10.33, 10.47 Ex. 11.9, 11.15, 11.24, 11.36 Ex.12.3, 12.10, 12.18 General Review for Chapters 10 to 12 Chapt. 13 Chapt. 14 May/10/10 Assignment Installation of software for labs; Mathematics test & review Ex.3.4, 3.9, 3.26 Ex. 4.4, 4.11, 4.13, 4.17, 4.28 Ex. 13.9, 13.11, 13.15, 13.20, Ex 14.2, 14.4, 14.12 Final Examination 10.00 to 11.50 NOTE: Homework is due on the date specified; e.g. on Feb. 12 exercises with numbers “6.xx” are due. Note: For some computer exercises, you may have to refer to previous exercises describing the computer steps in detail; e.g. to do Ex. 3.26 requires reference to Ex. 3.6 for plotting histograms and reference to Ex. 3.19 for multiple Box and Whisker plots. Project Topics For each of the suggested projects listed below you are urged to begin your evaluation of which topic to pursue by reading the relevant sections of Lomborg’s “The Skeptical Environmentalist.” The reason for beginning here is that Lomborg is a statistician who is trying to assess the validity of factual evidence and more importantly he cites data references extensively. Starting with a statistician’s assessment of the situation and a set of references including access to data is a useful way to begin your own investigation. Do try to discover your own data sources. You are also urged to extend your statistical analysis to procedures not yet studied in class. Your team may suggest a different topic, but have it approved by me first. Deforestation in the U.S., or in Europe, or in the World: -role of newsprint -acid rain Global temperature variation; -long term cycles or trends? Has there been a fall in male sperm counts? Is the world running out of water? What are the changes in obesity in the U.S. and in Europe? An analysis of the costs per life year saved for a selection of life saving measures. Cost benefit analysis of genetically modified food stocks in third world countries Is the incidence and severity of hurricanes increasing? -is the cost of hurricane damage increasing? Species extinction: best estimates and caveats. Cost benefit analysis of DDT: Latest evidence. OUTLINE FOR PAPER TITLE PAGE, page 0: Centered: Title of paper, author(s), {Paper for Analytical Statistics}, date and place. PAGE 1: Introduction: introduces the subject and indicates the precise objective of the research to be quoted. PAGE 2FF: Brief review of the relevant literature leading to a precise statement of what the authors intend to do and why; references cited, e.g. Jones(2001). PAGE 3FF: Statement of what calculations made and why followed by an evaluation of results obtained. Tables and graphs at very end of text, references cited. PAGE 4: Summary and Conclusions PAGE 5: Bibliography Entries to follow standard bibliographic practice; full references given, not just URL paths. PAGE 6FF: List of Tables and Graphs; each descriptively titled, graph axes labeled, and complete references to underlying data sources for all tables and graphs.