Course Objective: Understand deeply the theory and application of the linear regression model in economics. This is a first-year PhD course.
Reading: The textbook is Econometrics by Bruce Hansen, available free here.
Computing: We use R extensively in the class. Materials will be posted here soon.
Course Outline:
- Conditional Expectations and Projection (1 week)
- Reading: Hansen Ch 1-2; Wooldridge Ch 1-2; Greene Ch 1.
- Topics: Conditional expectations function; law of iterated expectations; linear projection; omitted variable bias; causal effects
- Least Squares Regression (2 weeks)
- Reading: Hansen Ch 3-4; Hayashi Ch 1.1-1.2, Greene Ch 2, 3, 4.1-4.3.
- Topics: Matrix notation; orthogonal projection; Frisch-Waugh-Lovell; mean and variance of OLS; Gauss-Markov theorem; GLS; covariance matrix estimation
- Small samples: Normal Regression (½ week)
- Reading: Hansen Ch 5; Hayashi Ch 1.3-1.7; Greene 4.3.
- Topics: Normal, chi-square, F, and t distributions; distribution of OLS coefficient vector; hypothesis tests and confidence intervals
- Asymptotic Theory for Least Squares (1½ weeks)
- Reading: Hansen Ch 6-7; Wooldridge Ch 3, 4.1-4.2; Hayashi Ch 2; Greene 4.4-4.6.
- Topics: Consistency; asymptotic normality; covariance matrix estimation; hypothesis tests and confidence intervals
- Hypothesis Testing (1 week)
- Reading: Hansen Ch 9; Greene Ch 5.
- Topics: Size and power; Wald and Hausman tests; multiple testing; local power; data mining; post model-selection inference
- The Bootstrap (1 week)
- Reading: Hansen Ch 10; Greene Ch 15.1-15.5.
- Topics: Nonparametric bootstrap; parametric bootstrap
- Endogeneity (1 week)
- Reading: Hansen Ch 12; Wooldridge Ch 5; Greene Ch 8.
- Topics: Instrumental variables and two stage least squares; weak instruments
- Time Series Analysis (1 weeks)
- Reading: Hansen Ch 14; Hayashi Ch 6.1-6.5, 9; Greene Ch 20-21.
- Topics: Stationarity and ergodicity; robust inference; spurious regression