Econometric Analysis of Panel Data

Class Notes

Professor William Greene
Department of Economics
Office:MEC 7-90, Ph. 998-0876
e-mail:wgreene@stern.nyu.edu
URL: http://people.stern.nyu.edu/wgreene

prev0.gifReturn to course home page.

Abstract: This is an intermediate level, Ph.D. course in the area of  Applied Econometrics dealing with Panel Data.  The range of topics covered in the course will span a large part of econometrics generally, though we are particularly interested in those techniques as they are adapted to the analysis of 'panel' or 'longitudinal' data sets.  Topics to be studied include specification, estimation, and inference in the context of models that include individual (firm, person, etc.) effects.  We will  begin with a development of the standard linear regression model, then extend it to panel data settings involving 'fixed' and 'random' effects.  The asymptotic distribution theory necessary for analysis of generalized linear and nonlinear models will be reviewed or developed as we proceed.. We will then turn to instrumental variables, maximum likelihood, generalized method of moments (GMM), and two step estimation methods.  The linear model will be extended to dynamic models and recently developed GMM and instrumental variables techniques.   The classical methods of maximum likelihood and GMM and Bayesian methods, expecially MCMC techniques, are applied to models with individual effects.  The last third of the course will focus on nonlinear models.  Theoretical developments will focus on heterogeneity in models including random parameter variation, latent class (finite mixture) and 'mixed' and hierarchical models.  We will also visit the theory for  techniques for optimization in the setting of nonlinear models.  We will consider numerous applications from the literature, including static and dynamic regression models, heterogeneous parameters models (e.g., Fama-Macbeth), random parameter variation, and specific nonlinear models such as binary and multinomial choice and models for count data. 

Notes: The following list points to the class discussion notes for Econometric Analysis of Panel Data. These are Powerpoint .pptx files.  

1. Introduction to Econometrics; Introduction to the course

2. Statistical Models: Estimation and Testing; The linear model

2-A. Endogeneity in the linear model

3. Models with Individual Effects

4. Fixed Effects and Hierarchical Models

4-A. Minimum Distance Estimation

5. Random Effects Models

6. Random Effects Model: Maximum Likelihood Estimation. Panel Data Structures

7. Extensions of Effects Models; Time Varying Fixed Effects, Heteroscedasticity, Measurement Error, Spatial Autocorrelation

8. Instrumental Variables; The Hausman-Taylor Estimator, GMM Estimation

9. GMM Estimation, Dynamic Models, Arellano/Bond/Bover, Schmidt and Ahn

10. Dynamic Models, Time Series, Panels and Nonstationary Data

11. Heterogeneous Parameter Models (Fixed and Random Effects), Two Step Analysis of Panel Data Models

12. Random Parameters, Discrete Random Parameter Variation, Continuous Parameter Variation

13. Linear Regression and Nonlinear Modeling

14. Nonlinear Models and Nonlinear Optimization; ML Estimation, M Estimation, GMM Estimation

15. Classical Estimation of Nonlinear Effects Models; Random and Fixed Effects Binary Choice Models

16. Random and Fixed Effects in Nonlinear Models, Quadrature and Simulation

17. Discrete Choice Models for Spatial Data

18. Ordered Choices and Censored Dependent Variables - Microeconometrics

19. Limited Dependent Variable Models and Models for Count Data

20. Sample Sample Selection Models and Models of Attrition

21. Hazard Function and Duration Models

22. Stochastic Frontiers and Efficiency Estimation, Applications from the Stochastic Frontiers Literature

23. Random Parameters Models, Heterogeneity, Second Generation, Simulation Based Estimation

24. Multinomial Choice and Stated Choice Experiments

24-A. Multinomial Choice Model Extensions: Best/Worst Data, Hybrid Choice

25. Bayesian Estimation Gibbs Sampling, Markov Chain Monte Carlo, Multinomial Choice, Economics and Marketing Application

26. Bayesian Methods and Models of Heterogeneity

FINAL EXAM (NO CLASS MEETING)    Click here to download the final examination.

prev0.gifReturn to course home page.