Econ 334: Econometrics

Econometrics connects economic theory to economic reality by translating the explanations, predictions, and testable implications of theory into statistical models which can be fit to economic data. We will focus on regression models and their many extensions, with an eye on what can go wrong with the models in various situations, and how to fix them when it does go wrong. Topics covered will include basic linear regression and extensions to probability models, intrumental variables, and time-series regressions.

The end goal of the course is that the student becomes a critical reader of others' work and a sophisticated producer of his or her own econometric models. Understanding the mathematics of regression analysis is absolutely necessary for both goals, and a fair amount of work on the underlying math is to be expected. At the same time, there will be a strong applied component to the course, and students will work with real-life economic datasets and current computer-based methods.

Instructor

TAs

Past Exams

[Midterm I. fall 2007]
[Midterm II. fall 2007]
[Final fall 2007].

Class 10: Review

Class 9: Panel Data

(Class Notes)

Class 8: Instrumental Variable Regression

(Class Notes)

(Practice Midterm)

Class 7: Internal and External Validity

(Class Notes)

Class 6 Multiple Regression: Hypothesis Tests

(Homework 3), due March 24.

(TeachingRatings data: EXCEL, format: .xls)

(TeachingRatings data: STATA, format: .zip)

(TeachingRatings data description)

The Midterm Exam:

You won't need anything but pencils, erasers, and a good cumulative knowledge of the material covered in the class.

Class 5: Multiple Regression: Inference

(Class Notes)

(Practice Midterm)

Class 4: Multiple Regression: Estimation

(Class Notes)

Class 3: Simple Linear Regression: Estimation and Interpretation

(Handout)

(Homework 2), due Feb 25.

Class 2: The Math and Logic of Simple Linear Regression

Class 1: Introduction

(Homework 1), due Feb 11.

Syllabus:

Textbook:

The course uses Stock and Watson "Introduction to Econometrics" as the only required text. It's a fantastic textbook, and is well worth reading and referring to later. At times we'll cover material in class from a perspective different from the book's, though, so reading the book alone will not be enough.

Sessions:

Sessions are mandatory and will cover a lot of the nitty-gritty of the course material. The sessions will focus more on problem-solving than the lectures, which many students will find to be extremely helpful. Attend session.

Homework:

Learning by doing is extremely important in econometrics courses, as with most math. Students who take the homework seriously should have little trouble with the exams.

Exams:

There will be two midterm exams and one final exam. The final will be cumulative.

Software:

You can do everything in Excel. However, more advanced programs are available if you would like to try.

The first one is STATA and the TA will give an introduction to STATA in the section. The second one is [R statistical programming language]. R is free, open-source, easy to install, runs on nearly any computer, and makes a lot of the statistical work we'll be doing very easy.