Economics 696: Topics in EconometricsSpring Semester 2008Course SyllabusLectures: Wednesdays 12:30-3:00 pm, McClelland 401KKInstructor: Keisuke Hiranoemail: [email protected] Hours: Wednesdays, 10-11:30 am, or by appointment.Teaching Assistant: Dong-Hyuk Kimemail: [email protected] Description: This topics course will focus on decision-theoretic metho ds for analyzingdata and making policy prescriptions. We will look both frequentist and s ubjectivist formulationsof the basic statistical decision problem, and compare them. We will also study mo dern Bayesianinference methods, which often require numerical methods including Markov chain Monte Carlo,to simulate posterior distributions and related quantities, and also frequentist simulation-basedestimation procedures, such as the simulated me thod of moments, and discuss both practical im-plementation and asymptotic theory for such methods. If time permits we will discuss Bayesiannonparametrics and semiparametrics.Prerequisites: Economics 520, 522A, and 522B, or equivalent. You are expected to be familiarwith basic probability and statistics, linear regression theory, instrumental variables and methodof moments/M-estimator theory, and conditional maximum likelihood techniques for nonlinear re-gression models. You also need to have familiarity with programming in a standard language sothat you can write code to implement estimators and other procedures for the course.Textbook: Most of the course will b e based on lecture notes and articles. There is no requiredtextbook for the course, but some readings will be drawn from the following bo ok:• Geweke, J., 2005, Contemporary Bayesian Econometrics and Statistics, Wiley.Other good references for decision theory and Bayesian analysis include:• Berger, J., 1985, , Statistical Decision Theory and Bayesian Analysis, 2nd ed., Springer-Verlag.• Carlin, B., and Louis, T., 2000, Bayes and Empirical Bayes Methods for Data Analysis, 2nded., Chapman and Hall/CRC Press.• Ferguson, T., 1967, Mathematical Statistics: A Decision Theoretic Approach, Academic Press.• Gelman, A., Carlin, J., Stern, H., and Rubin, D., 2003, Bayesian Data Analysis, 2nd ed.,Chapman and Hall/CRC Press.• Lancaster, T., 2004, An Introduction to Modern Bayesian Econometrics, Blackwell.1Assessment: There will be a small number of problem sets, involving a combination of analyticwork and data analysis. There will also be a final project in which you choose an empiricalpaper to reexamine using methods developed in the course. You may use any standard statisticalpackage/programming language you wish. I recommend either Matlab or R.Course Web Site: http://www.u.arizona.edu/~hirano/696 2008.htmlI will post lecture notes, homework assignments, and other supplementary material for the coursehere.Outline: (may be revised as semester progresses)1. Introduction: Statistical Decision Theory and Bayes ProceduresLecture NotesBarberis, N., 2000, “Investing for the Long Run when Returns are Predictable,” Journal ofFinance 55, 225-264.Dehejia, R., 2005, “Program Evaluation as a Decision Problem,” Journal of Econometrics125, 141-173.Efron, B., 2005, “Bayesians, Frequentists, and Scientists,” Journal of the American StatisticalAssociation 100(469), pp. 1-5.Geweke, Ch. 1.2. Foundations of Statistical Decision Theory and Bayesian InferenceLecture NotesFerguson, Ch. 1-2.3. Normal-based modelsLecture NotesGeweke, Ch. 2.4. Bayesian computationsLecture NotesAlbert, J. H., and S. Chib, (1993), “Bayesian Analysis of Binary and Polychotomous ResponseData,” Journal of the American Statistical Association, Vol 88, No. 422, pp. 669-679.*Chib, S., and E. Greenberg, (1996), “Markov Chain Monte Carlo Simulation Methods inEconometrics,” Econometric Theory, 12(3), 409-431.Geweke, Ch. 4.Carlin and Louis, Ch. 5.Gelman, Carlin, Stern, Rubin, Ch. 10,11.5. Hierarchical and Panel ModelsLecture NotesGelman, Carlin, Stern, and Rubin, Ch.5.Rossi, P., R. McCulloch, and G. Allenby, (1995), “Hierarchical Modelling of Consumer Hetero-geneity: An Application to Target Marketing,” in Case Studies in Bayesian S tatistics, Vol. II,Lecture Notes in Statistics 105, ed. C. Gatsonis, J. Hodges, R. Kass, and N. Singpurwalla,New York: Springer–Verlag.*2Geweke, J., Gowrisankaran, G., and Town, R., 2003, “Bayesian inference for hospital qualityin a selection model,” Econometrica 71, 1215-1238.6. Simulation-based method of moments: introductionMcFadden, D., 1989, “A Method of Simulated Moments for Estimation of Discrete ResponseModels without Numerican Integration,” Econometrica 57, 995-1026.Pakes, A., and Pollard, D., 1989, “Simulation and the Asymptotics of Optimization Estima-tors,” Econometrica 57, 1027-1057.Hajivassiliou, V., and McFadden, D., 1998, “The Method of Simulated Scores for Estimationof LDV Models,” Econometrica 66(4), 863-896.7. Empirical Process Theory and Asymptotics for Simulation-based estimatorsLecture NotesPakes, A., and Pollard, D., 1989, “Simulation and the Asymptotics of Optimization Estima-tors,” Econometrica 57, 1027-1057.8. Treatment Assignment and Testing: frequentist and Bayesian approachesDehejia, R., 2005, “Program Evaluation as a Decision Problem,” Journal of Econometrics125, 141-173.Hirano, K., and Porter, J. R., 2005, “Asymptotics for Statistical Treatment Rules,” workingpaper, University of Arizona.Manski, C. F., 2002, “Treatment Choice Under Ambiguity Induced by Inferential Problems,”Journal of Statistical Planning and Inference 105, 67-82.Manski, C. F., 2004, “Statistical Treatment Rules for Heterogeneous Populations,” Econo-metrica 72, 1221-1246.9. Nonparametrics and Semiparametrics: decision-theoretic approachesFerguson, T. S., 1974, “Prior Distributions on Spaces of Probability Measures,” The Annalsof Statistics 2, 615-629.Escobar, M., and M. West, 1995, “Bayesian Density Estimation and Inference using Mix-tures,” Journal of the American Statistical Association 90,