Course Syllabus

Econ 527
Econometric Methods
2020-21 Winter Session

Course description:

This course is a rigorous introduction to econometric theory. The static linear regression model is the main focus of the course. We will cover estimation and testing methods based on ordinary least squares (OLS), generalized least squares (GLS), generalized method of moments (GMM) and instrumental variables (IV). We will also discuss maximum likelihood (ML) estimation of binary choice models (logit and probit). 
Students are assumed to be familiar with the basic concepts of linear algebra, multivariate calculus and statistics. The requisite linear algebra and statistics results will be reviewed at the beginning of the course. A concise but excellent treatment of topics from basic probability to statistical inference can be found in Introduction to Statistics and Econometrics by T. Amemiya. A brief review of linear algebra and probability theory can be found in Appendices A and B of Econometric Analysis by W. H. Greene.
There is no required textbook. All the required material will covered in my lecture notes and/or iPad notes from the online lectures.
The following books are recommended for students interested in additional resources:
  • Davidson, R., and J. G. MacKinnon (2004): Econometric Theory and Methods, Oxford University Press, New York. This excellent textbook is particularly recommended for its treatment of the geometric properties of OLS and other methods (see topics 1, 2, 5, and 6 below).
  • Greene, W.H. (2017 or earlier editions): Econometric Analysis, Pearson. This is one of the most comprehensive graduate textbooks on econometric theory. Make sure to check Appendix A if you need to refresh your matrix algebra.

Tutorials:

  •  Wed, 8:30-10AM, Buch B313

To speed up our progress, discussion of some of the material from the lecture notes will be deferred to the tutorials.

Assignments:

There will be weekly problem sets including analytical and practical questions. Practical questions will require handling and analyzing data using R. R is a free, powerful, and extremely popular statistical software. In addition to R, it is also recommended to install RStudio, also a free and popular software that provides a convenient interface for R. Computer assignments should be submitted by uploading R Markdown (RMD) and output HTML files to Canvas (the detailed instructions will be given at a later time).
Students should expect and prepare for practical questions with R on the midterm and final exams.

R:

Some basic R training will be provided at the beginning of the semester. There are also numerous resources on R available online, see for example:

Grading:

  • Assignments: 10%
  • Midterm: 35%
  • Final: 55%

Topics: 

  1. Linear regression model. Multiple linear regression, OLS and its small sample properties, geometry of LS, partitioned regression, goodness of fit.
  2. Hypothesis testing and confidence intervals (small sample).
  3. Large sample theory. Convergence in probability, convergence in distribution, delta method, laws of large numbers, central limit theorems.
  4. Large sample regression theory. Consistency and asymptotic normality of the LS, hypothesis testing and confidence intervals.
  5. GLS.
  6. Instrumental Variables estimation.
  7. GMM.
  8. Simultaneous equations.
  9. Maximum Likelihood.
  10. Binary choice models.
  11. Time series methods.