EC827 Econometric Theory I Dep't of Economics Prof. Arthur Lewbel Spring 2007 Boston College room 491, 552-3678 OVERVIEW The goal of this course is to provide an understanding of the econometric theory that underlies common econometric models. The focus is on the regression model and its many extensions. Portions of the course will cover topics students have seen before, but these will be covered in greater depth. After taking this course, students should be able to: 1. Choose appropriate models and estimators for given economic applications. 2. Interpret model estimates. 3. Diagnose potential problems with models and know how to remedy them. 4. Possess a sufficient grounding in econometric theory to begin advanced work in the field. The textbook for this course is Greene, W. H., "Econometric Analysis," 5th edition, Prentice Hall, plus a few additional readings. The following syllabus give 5th edition chapters in Greene. SYLLABUS 1. Regression vs correlation and causes. 2. Finite Sample Properties of Estimators. G Appendix C 3. Asymptotic Properties of Estimators. G Appendix D, Also Eric Zivot's Primer on Asymptotics, available at http://www2.bc.edu/~lewbel/828zivot.pdf 4. Classical regression. G 4,5. 5. Specification issues: multicollinearity, coefficient interpretation, dummies. G 7,8. 6. Maximum likelihood estimation G 17 7. Inference, hypothesis tests. G 6. 8. GLS, non-iid errors (autocorrelation and heteroskedasticity) G 10,11,12. 9. Dynamic and time series models G 19,20. 10. IV and 2SLS estimation, endogeneity and simultaneity. G 5,14,15. 11. Nonlinear models and GMM (G 9,10), Possible additional topics, time permitting: Extemum estimators, discrete dependent variable models (G 21,22), Panel Data models (G 13). Other books you may find useful for additional reading are: Amemiya, T. (1985) Advanced Econometrics. Spanos, A., (1990) Statistical Foundations of Econometric Modeling. Mittelhammer, R.C., G.G. Judge, and D.J. Miller, (2000) Econometric Foundations. Wooldridge, J. M. (2002) Econometric Analysis of Cross Section and Panel Data. GRADING: midterm: 50%, Final: 50%.