Advanced Econometrics II

Course Syllabus (Chinese Version)

Lectures at Xiamen University, China
Part I: December 13-19, 2005. Part II: March 29 - April 28, 2006
http://www.econ.pdx.edu/staff/KPL/XMU/
Prof. Kuan-Pin Lin
Portland State University
Portland, Oregon 97207, USA

This 40-hours course covers advanced topics related to methodological issues in econometrics, with emphases on computation intensive methods and applications. This course is divided into two parts. The first part (10-hours) is a preparatory course introducing econometric computing with GAUSS and reviewing the linear econometric models and applications. The second part (30-hours) covers non-linear regression models, time series models, and qualitative choice models. The purpose of this course is to prepare students with broad knowledge of econometric methods and applications capable of doing independent research project. In addition to economic theory, knowledge of basic econometrics is required. Experience of computer programming is helpful but not necessary. GAUSS and GPE2 econometric package will be used throughout the course.

Texts and References

• W. H. Greene, Econometric Analysis, 5th ed., Prentice Hall, 2002.
• F. Hayashi, Econometrics, Princeton University Press, 2000.
• K.-P. Lin, Computational Econometrics: GAUSS Programming for Econometricians and Financial Analysts, ETEXT Publishing, 2001; Chinese Edition, Tsinghua University Press, 2003.
• J. M. Wooldridge, Econometric Analysis of Cross Section and Panel Data, The MIT Press, 2002.

Software and Manual

• GAUSS 7.0 for Windows Student Version, Aptech Systems, 2005. Academic and professional versions of GAUSS for Windows are also available from Aptech.
• GPE2 (GAUSS Programming for Econometricians and Financial Analysts) for GAUSS, included in Computational Econometrics. Free upgrade of GPE2 for GAUSS 7.0 is available here.

Course Topics

Econometric Computing with GAUSS Introduction to GAUSS Using GPE2 for GAUSS	Linear Econometric Models: Review and Applications Linear Restrictions and Structural Change Heteroscedasticity and Autocorrelation Instrumental Variables and Dynamic Models
Nonlinear Optimization Unconstrained Optimization Constrained Optimization	Nonlinear Regression Models Nonlinear Least Squares (NLSQ) Maximum Log-Likelihood (ML) Statistical Inferences in Nonlinear Models
Nonlinear Econometric Models Box-Cox Variable Transformation Hetroscedastic Regression Models Regime Switching Regression Models	Generalized Method of Moments Nonlinear Generalized Method of Moments (GMM) GMM Estimation for Econometric Models Application: A Nonlinear Rational Expectation Model
Time Series Analysis Nonstationary Time Series: Unit Roots and Cointegration Autoregressive and Moving Average (AR, MA, ARMA) Advanced Topics (ARMAX, GARCH, VAR, State-Space)	Qualitative Choice Models Binary and Multinomial Choice Models: Probit, Logit Limited Dependent Variable Model: Tobit Count Data and Posisson Model Duration Data and Hazard Function

Case Studies

• Chinese Agricultural Production Function
• Chinese Money Demand Equation
• Consumption-Income Relationship in China (Time Series)
• Consumption-Income Relationship in China (Panel Data)
• Real GDP and GDP Growth in China

Course Expectation

This course consists of lectures, readings, homework, project and final exam. Doing homework using GAUSS is important for not only to understand the theoretical concepts but also to learn structural and efficient computing techniques for econometric estimation and inference.

Guideline on Writing a Course Project

Format

1. 5 pages typed (double-space and wide margins, in English or Chinese).
2. The model presented has to be an original econometric model.
3. The format of the paper should follow a standard journal article closely.
4. Supporting data and computer program printout have to be included, but not counted for the page number.

Contents

1. Introduction and brief discussion of the main results.
2. Full explanation of model specification, estimation, and hypothesis testings.
3. Detailed interpretation of the model and its policy implication, if any.
4. Conclusion and future extensions.
5. References (including data sources).

Deadline

• The topic is expected to be a study of linear econometric model, but not limited to the Chinese economy.
• The project is evaluated based on its originality, creativity, and consistency with the format and content requirements described above.
• Project due: March 28, 2006 (before Part II of lectures begin on March 29).