ECONOMETRICS II

 

 

 

Fall 2006                                                                                 

ECON 629/AREC 634

MW 9:00-10:15                                                                      

Saunders 637

Professor SH Lee


COURSE OUTLINE

 

Course Description

 

This is the second part of course exploring econometric techniques used by research economists and other policy analysts. Main topics include single equation models under non-ideal conditions, simultaneous equation models, panel data models, maximum likelihood estimation, generalized methods of moments, discrete choice models, limited dependent variable models, stationary and non-stationary time series models, cointegration, vector autoregression, and forecasting.

 

The prerequisite for this course is ECON 628/AREC 626. A good understanding of the linear regression model is a must.

 

Required Text

 

            Wooldridge, Jeffrey. M. 2002. Econometric Analysis of Cross Section and Panel Data, MIT Press.

             

Optional Texts

 

            Hayashi, Fumio. 2000. Econometrics, Princeton University Press.

            Greene, William H. 2002. Econometric Analysis, 5th ed. Prentice Hall.

            Kennedy, Peter. 2003. A Guide to Econometrics, 5th ed. MIT Press.

            Maddala, G.S. 1983. Limited Dependent and Qualitative Variables in Econometrics, Cambridge University Press.

            Hamilton, James D. 1994. Time Series Analysis, Princeton University Press.

            Wooldridge, Jeffrey M. 2006. Introductory Econometrics: A Modern Approach, 3rd ed. Thomson/Southwestern

 

Hayashi and Greene are comprehensive econometrics textbooks.  Kennedy is good for overview and concepts. Maddala is the standard reference for limited dependent variable models. Hamilton is a comprehensive reference for time series models. Some articles will also be handed out in class. For quick review of basic econometrics, see Wooldridge.

 

 

COURSE REQUIREMENTS

 

There are four requirements. First, students must complete one empirical project. You will be responsible for defining a problem of interest, collecting data or finding an appropriate data set, developing an appropriate model, and evaluating it using the techniques which we will develop in the course. The project will be due the last day of classes.  A proposal for the project will be due in class, November 9 (Thursday), which will include a definition of the problem you are interested in investigating and a relevant data set. Second, homework will be assigned and to be turned in on a regular basis. Homework assignments are due at the beginning of the class. No late assignments will be accepted. Homework problems will involve both theoretical computations and practical applications which will need to be done using a statistical software package. Third, students must take the midterm exam. Four, students must take the final exam. The final exam is comprehensive. Exams will be a closed book and closed note, but I will hand out a cheat-sheet. In the event that an emergency arises, you are responsible to contact me before the exams to make alternative arrangements.

 

            Midterm Exam, Monday, October 16, 9:00-10:15 AM               30%

            Final Exam, Monday, December 15, 7:30-9:30 AM                   40%

            Homework                                                                                 20%

            Empirical Project                                                                        10%

                         

HELP and FACILITY

 

Help:

Office Hours: MW 10:30-11:50

Office: Saunders 512

Phone: 956-8590

E-mail: leesang@hawaii.edu

Web: www2.hawaii.edu/~leesang/

 

Statistical Software:

The problem sets will require you to familiar with statistical package programs such as STATA, SAS, SPSS, SHAZAM, LIMDEP or GAUSS. You may use any software, but I will use STATA which is available for students to use in the computer lab. I found STATA, GAUSS, and LIMDEP are all very good with some advantages and disadvantages. An introduction to the STATA will be given, but it is expected that you spend some time learning the capabilities of the software on your own or in groups. STATA includes tutorial programs that are particularly good for duration models. Computer labs in Saunders Hall have STATA. For account information visit http://www.socialsciences.hawaii.edu/pages/tech/lab/lab_account.html

Considerable information about these and other related programs and the latest techniques can also be found from their websites:

 

STATA is www.stata.com GAUSS is www.aptech.com LIMDEP is www.limdep.com
 

COURSE SCHEDULE

 

0. Carrying out an Empirical Project

1. Review of Single Equation Linear Models

2. Generalized Methods of Moments

3. Estimation by Two Stage Least Squares

4. Estimation of Systems of Equation Models

5. Panel Data Models

6. Maximum Likelihood Estimation

7. Models for Discrete Choice: Binary, Mutinomial, and Ordered Choice Models

8. Limited Dependent Variable and Duration Models

9. Stationary and Non-stationary Time Series Models and Cointegration

10. Vector Autoregressions

11. Forecasting