Скачать 30.12 Kb.

Marc Nerlove Revised: 11/05/05 AREC699, Fall 2005: Special Topics in Econometrics, Study Group Dates, Times, Topics and Readings, and Presenters Rules of the game: Each student, whether registered for independent study with me or AREC669J with Professor Just, who wishes to attend our study group, must attend the plenary session on Wednesday, September 14, 3:305:30, Room 2200B, to assume responsibility for presenting all or part of one of the following topics. Thereafter, each must attend every one of the 7 2hour sessions on each of the topics, do the suggested reading, and participate. If there is not a quorum of at least 6 students, the session will be cancelled. Failure to attend the plenary session will mean the student will not be allowed to attend. If you fail to attend other sessions, I may throw you out We will meet at 3:30 in Room 2200B Symons, unless Nancy is meeting her class that day. Day when Nancy meets, we ill meet at 3:00. Such days are marked with an asterisk. Those responsible for presenting each topic to the group are indicated. Topics, some suggested readings and suggestions for further reading: You should not panic at the length of this reading list. Suggested readings are pretty basic, but those responsible for particular sessions may want to suggest what they would like the group to focus on. Further suggestions are just that: some references for anyone who wants to go deeper into a topic. Many readings in both categories are available on the internet. I have not verified whether all of the books are available in McKeldin or the Engineering Library, but I suppose they are. Prior to the first topical session, everyone should review basic probability and statistics as covered in Martinez and Martinez, Computational Statistics with Matlab, Chaps. 13, pp. pp. 178. To rejuvenate your Matlab, I suggest the following selection of exercises; 2.12.5, 2.112.19; 3.13.9, 3.43.16. Those who take responsibility for each of the 7 topics, should email our group prior to that session to emphasize which readings will be especially relevant to their presentations. My draft of the syllabus is preliminary. I may suggest some further readings in the course of the semester or remove some, but here are my suggestions for now listed in rough order of importance: September 28, 3:30, Topic 1: Random Number Generation and Descriptive Statistics Presented by Mafuz and Dave Suggested Readings: Nerlove, Notes, 1998: "Random Number Generation," Sec. 1 at http://www.arec.umd.edu/montecarlo/MonteCarlo1.doc Cleve Moler, Fall 1995, "Random Thoughts," at http://www.mathworks.com/company/newsletters/news_notes/pdf/Cleve.pdf Cleve Moler, Spring 2001, "Normal Behavior," at http://www.mathworks.com/company/newsletters/news_notes/clevescorner/spring01_cleve.html Martinez & Martinez, Chapter 4, "Generating Random Variables," pp. 79110. I suggest the following selection of exercises: 4.14.2, 4.44.9, 4.11412, 4.14. Martinez & Martinez, Chapter 5, "Exploratory Data Analysis," Sections 5.1 5.3,pp. 111146. I suggest the following exercises: 5.15.2. Further suggestions: Press, W. H.; Flannery, B. P.; Teukolsky, S. A.; and Vetterling, W. T. "Random Numbers." Ch. 7 in Numerical Recipes in FORTRAN: The Art of Scientific Computing, 2nd ed. Cambridge, England: Cambridge University Press, pp. 266306, 1992. You can access the entire book at http://www.library.cornell.edu/nr/cbookcpdf.html . Don't be misled by the use of FORTRAN in the title. The exposition is superb and does not require knowledge of FORTRAN to follow. The following articles in Mathworld, http://mathworld.wolfram.com/ , are worth checking out: Hammersley Point Set, Pseudorandom Number, Quasirandom Number, Random Number, Uniform Distribution Theory, van der Corput Sequence. Quasirandom numbers or low discrepancy sequences may be better than the pseudo random numbers which can be generated directly in Matlab. There are some thirdparty programs which I can make available. Mathematica actually has an option to do numerical Monte Carlo integration using such sequences. Roman Maeder has a Mathematica package, which will generate lowdiscrepancy sequences at http://library.wolfram.co.jp/conferences/conference98/abstracts/quasi_random_numbers.html . You can get a free demo at: http://www.mathdirect.com/products/qrn/ . You can access the following via UM Research Port; "Random number generation for the new century," LihYuan Deng; Dennis K J Lin, The American Statistician; May 2000; 54, 2 "Pseudorandom Numbers," Jeffrey C. Lagarias, Statistical Science, Vol. 8, No. 1, Report from the Committee on Applied and Theoretical Statistics of the National Research Council on Probability and Algorithms. (Feb., 1993), pp. 3139 *October 26, 3:00pm,Topic 2: Monte Carlo Methods for Statistical Inference Presented by Constant and Michele Suggested Readings: Nerlove, Notes, 1997, Sec. 2, "Monte Carlo Methods," at http://www.arec.umd.edu/montecarlo/MonteCarlo2.doc . Martinez & Martinez, Chapter 6, Sections 6.1  6.3, pp. 191214. I suggest the following exercises: 6.2, 6.4, 6.86.9, 6.12. Martinez & Martinez, Chapter 11, "Markov Chain Monte Carlo Methods," pp. 425464. I suggest the following exercises: 11.1  11.5, 11.10  11.11. Gentle, Elements of Computational Statistics, Springer, 2002, Chapter 2, "Monte Carlo Methods for Statistical Inference," pp. 3968. Appendix A, "Monte Carlo Studies in Statistics," pp. 337350. Further Suggestions: Hendry, D. F., "Monte Carlo Experimentation in Econometrics," Chapter 16, pp. 937976, in Z. Griliches and M.D. Intriligator, Handbook of Econometrics, Vol. II, Amsterdam: Elsevier, 1984. I can ask Jane to make a copy of this, if it is not available in the AREC Library. Givens & Hoeting, Computational Statistics, Wiley, 2005, Chapter 5, "Numerical Integration," pp. 121142; Chapter 6, "Simulation and Monte Carlo Integration," pp. 143182; Chapter 7, "Markov Chain Monte Carlo," pp.183218. Geweke, J., and M. Keane, "Computationally Intensive Methods for Integration in Econometrics, 34633568 in Handbook of Econometrics, Vol.5 ed. by J. J Heckman and Edward Leamer, Amsterdam: Elsevier Science, 2001. I can ask Jane to make a copy of this for the group, if it is not available in the AREC Library. Chib, S., "Markov Chain Monte Carlo Methods: Computation and Inference, pp. 3578 3649, in Handbook of Econometrics, Vol.5 ed. by J. J Heckman and Edward Leamer, Amsterdam: Elsevier Science, 2001. I can ask Jane to make a copy of this for the group, if it is not available in the AREC Library. Liesenfeld, R., and JF. Richard, "Simulation Techniques for Panels: Efficient Importance Sampling," unpublished. From The Econometrics of Panel Data: Fundamentals and Recent Developments in Theory and Practice, edited by L Mátyás and P Sevestre. Jank, Wolfgang, "Stochastic Variants of EM: Monte Carlo, QuasiMonte Carlo and More," Unpublished, available at: http://www.smith.umd.edu/faculty/wjank/StochasticVariantsOfEMProceedingsJSM2005.pdf You can access these articles via UM Research Port: "The Reporting of ComputationBased Results in Statistics," David C. Hoaglin; David F. Andrews, The American Statistician, Vol. 29, No. 3. (Aug., 1975), pp. 122126. "Understanding the MetropolisHastings Algorithm," Siddhartha Chib; Edward Greenberg, The American Statistician, Vol. 49, No. 4. (Nov., 1995), pp. 327335. "A History of the MetropolisHastings Algorithm," David B Hitchcock. The American Statistician, Nov 2003.Vol.57, Iss. 4; pg. 254258. "Explaining the Gibbs Sampler," George Casella; Edward I. George, The American Statistician, Vol. 46, No. 3. (Aug., 1992), pp. 167174. "Stochastic Simulation in the Nineteenth Century," Stephen M. Stigler Statistical Science, Vol. 6, No. 1. (Feb., 1991), pp. 8997. "Sharpening Buffon's Needle," Michael D. Perlman; Michael J. Wichura The American Statistician, Vol. 29, No. 4. (Nov., 1975), pp. 157163. *November 16, 3:00pm, Topic 3: Bootstrapping, Jackknifing, and Data Partitioning Presented by Marcela and Mike Suggested Readings: Nerlove, Notes, 1998, Section 3, "Bootstrapping," at http://www.arec.umd.edu/montecarlo/MonteCarlo3.doc Martinez & Martinez, Chapter 6.4, pp. 214226. Chapter 7, pp. 231258. I suggest the following exercises: 6.13, 7.47.7, 7.9. Gentle, Chapter 3, "Randomization and Data Partitioning," pp. 6984; Chapter 4, "Bootstrap Methods," pp. 8598. Givens and Hoeting, Chapter 9, "Bootstrapping," pp. 253275. Further Suggestions: Holmes, Susan P., Course Lectures and Other Notes for "Introduction to the Bootstrap." Unpublished, available at: http://wwwstat.stanford.edu/~susan/courses/s208/ Efron, B., and R. J. Tibshirani, An Introduction to the Bootstrap, Chapman & Hall, 1993. Hall, P., "Methodology and Theory for the Bootstrap," pp. 2341  2381 in Handbook of Econometrics, Vol. 4, ed. by R. F. Engle and D. L. McFadden, Amsterdam: Elsevier Science, 1994. I can ask Jane to make a copy of this for the group, if it is not available in the AREC Library. Horowitz, J. L., "The Bootstrap," pp. 31593228 in Handbook of Econometrics, Vol.5 ed. by J. J Heckman and Edward Leamer, Amsterdam: Elsevier Science, 2001. I can ask Jane to make a copy of this for the group, if it is not available in the AREC Library. You can access the following articles via UM Research Port: "A Leisurely Look at the Bootstrap, the Jackknife, and CrossValidation," Bradley Efron; Gail Gong, The American Statistician, Vol. 37, No. 1. (Feb., 1983), pp. 3648. "Bootstrap: More than a Stab in the Dark?," G. Alastair Young, Statistical Science, Vol. 9, No. 3. (Aug., 1994), pp. 382395. "Monte Carlo Approximation of Bootstrap Variances," James G. Booth; Somnath Sarkar, The American Statistician, Vol. 52, No. 4. (Nov., 1998), pp. 354357. November 23, 1pm  3pm,Topic 4: Optimization Note: In deference to those who would like to get away for the Thanksgiving Holiday, we will start at 1pm. Presented by Chad and Craig Suggested Readings: Go through the Tutorial for the Matlab Optimization Toolbox and the following demos: Largescale unconstrained nonlinear optimization Largescale constrained linear leastsquares Minimization of the "banana function" P. Venkataraman, Applied Optimization with Matlab Programming, Wiley, 2002, Chapter 5, "Numerical Techniques  The OneDimensional Problem," pp. 203226; Chapter 6, "Numerical Techniques for Unconstrained Optimization," pp. 227264; Chapter 7, "Numerical Techniques for Constrained Optimization," pp. 265317; Chapter 9, "Global Optimization," pp350378. I have asked Katherine to order a copy of this book for the AREC Library. There is a website which gives a link to all the Matlab code for this book: http://www.wiley.com/legacy/products/subject/engineering/venkat/ . Eric W. Weisstein et al. "Global Optimization." From MathWorldA Wolfram Web Resource. http://mathworld.wolfram.com/GlobalOptimization.html Further Suggestions: Givens & Hoeting, Chapter 2, "Optimization and Solving Nonlinear Equations," pp. 1948. Press, W. H.; Flannery, B. P.; Teukolsky, S. A.; and Vetterling, W. T., Chapter 9, "Root Finding and Nonlinear Sets of Equations"; Chapter 10, "Maximization or Minimization of Functions." You can access the entire book at http://www.library.cornell.edu/nr/cbookcpdf.html . November 30, 3:30pm, Topic 5: Likelihood and Maximum Likelihood Presented by Nate and Tim Suggested Readings: Nerlove, Likelihood Inference in Econometrics, Chapter 1, unpublished, 1999, "The Likelihood Principle," at http://www.arec.umd.edu/mnerlove/images/chapter1.pdf . Greene, William H., Econometric Analysis, any recent edition, the Chapters on "Maximum Likelihood Estimation," and on "Generalized Method of Moments; for example, Chapters 17 & 18 in the 5th edition. Further Suggestions: Edwards, A. W. F., Likelihood, Expanded Edition, Johns Hopkins Press, 1992. Royall, R., Statistical Evidence: A Likelihood Paradigm, London: Chapman & Hall, 1997. Severini, T. A., Likelihood Methods in Statistics, Oxford, 2000. Tanner, M. A., Tools for Statistical Inference: Methods for the Exploration of Posterior Distributions and Likelihood Functions, Third Edition, Springer, 1996. You can access these articles via the UM Research Port: "Laplace's 1774 Memoir on Inverse Probability," Stephen M. Stigler," Statistical Science, Vol. 1, No. 3. (Aug., 1986), pp. 359363. "A Note on Maximum Likelihood Estimation," Nicolas W. Hengartner, The American Statistician, Vol. 53, No. 2. (May, 1999), pp. 123125. "On the History of Maximum Likelihood in Relation to Inverse Probability and Least Squares," Anders Hald, Statistical Science, Vol. 14, No. 2. (May, 1999), pp. 214222. "Likelihood Inference and Time Series," G. A. Barnard; G. M. Jenkins; C. B. Winsten, Journal of the Royal Statistical Society. Series A (General), Vol. 125, No. 3. (1962), pp. 321372. *December 7, 3pm, Topic 6: The EM Algorithm Presented by Shinsuke (Yagi) and Jia Givens & Hoeting, Chapter 4, "EM Optimization Methods," pp. 89120. Wolfgang Jank, Stochastic Variants of EM: Monte Carlo, QuasiMonte Carlo, and More, at http://www.smith.umd.edu/faculty/wjank/StochasticVariantsOfEMProceedingsJSM2005.pdf Further Suggestions: McLachlan, G. J., and T. Krishnan, The EM Algorithm and Extensions, Wiley, 1997. You can access these articles via the UM Research Port "A Graphical Illustration of the EM Algorithm," William Navidi, The American Statistician, Vol. 51, No. 1. (Feb., 1997), pp. 2931. December 13, 3:30pm, Topic 7: Probability Density Estimation Presented by Anmol and Dimitrios Nerlove, 1999, unpublished, "Notes on Method of Moments and Mixture Distributions<" at: To be supplied. Martinez & Martinez, Chapter 8, "Probability Density Estimation," pp. 259316. I suggest the following exercises: 8.18.3, 8.88.9, and 8.18. Gentle, Chapter 8, "Estimation of Probability Density Functions Using Parametric Models," pp. 197203; Chapter 9, Nonparametric Estimation of Probability Density Functions," pp. 205231. Further Suggestions: Givens & Hoeting, Chapter 10, "Nonparametric Density Estimation," pp. 277311. Silverman, B. W., Density Estimation for Statistics and Data Analysis, Chapman & Hall, 1998. McLachlan, G., and D. Peel, Finite Mixture Models, Wiley, 2000. You can access these articles via the UM Research Port: "Difficulties in Drawing Inferences With FiniteMixture Models: A Simple Example With a Simple Solution," Hwan Chung, Eric Loken, Joseph L Schafer. The American Statistician, May 2004.Vol.58, Iss. 2; pg. 152159. 