Regression Analysis and Causal Inference


Course Description:
This course covers the Econometrics topics related to the linear regression models. We will discuss the estimation and identification of the linear regression models. We further explore the common mistakes and issues in the estimation, the corresponding statistical inferences, and the interpretation of the estimation results. The topics include the Gauss Markov Theorem, the Frisch–Waugh–Lovell theorem, regressor choices, endogenous regressors, non-spherical disturbances, measurement errors, and limited dependent variable models.

– Sample Mean and the Concept of Averages
– Linear Regression Basics and Gauss Markov Theorem
– Frisch–Waugh–Lovell theorem
– Omitted Variable Bias
– Bad Controls
– Instrumental Variables and Two-stage Least Squares
– Local Average Treatment Effects
– Fixed Effects and Random Effects
– Differences in Differences
– Regression Discontinuity and Regression Kink
– Generalized Least Squares and Feasible Generalized Least Squares
– Robust Standard Errors and Clustered Standard Errors
– Bootstrapping
– Measurement Errors
– Quantile Regression
– Linear Probability Model, Logit and Probit


February 25th~28th, 2019

13:30 ~ 17:00


Chungsang Tom Lam, The University of Chicago, Ph.D in Economics


Room 104, School of Information Management & Engineering, Shanghai University of Finance & Economics

Edit This Entry