2022 Intermediate Econometrics
Course Objective:
1. REVIEW: To help those who feel confused about the complicated proofs in EC310
2. UNDERSTAND: To illustrate the "over-mathematically-expressed" theories in plain language
3. IMPROVE: To explain some "advanced" knowledge beyond EC310
Wish you a pleasant journey with econometrics!
Note:
1. DO NOT SPREAD beyond "安心学"!!!!!!
2. VIDEO will be uploaded right after each class
3. SLIDES and CODE would be updated occasionally
4. In preparation of the contents, I borrow heavily from the lecture notes of Professor Xi Qu at Shanghai Jiao Tong University, who taught me introductory eonometrics. All errors are mine.
Videos:
(Click to watch)
Lecture 1: Simple Linear Model
Lecture 2~5: Multiple Linear Model
Lec 2-1: Model Setup; Lec 2-2: Stata Introduction;
Lec 3: T-Test & Matrix Representation Theorems;
Lec-4: F-Test & Projection Matrix & Matrix Form Test;
Lec 5: Asyptotic Theory & Data Generation Process
Lecture 8~9: Endogeneity
Lec 8: Soure of Endogeneity & Properties of IV Estimator;
Lec 9: Two Stage Least Squares; Applications of IV
Lecture 9~10: Time Series
Lec 9: Properties & Examples; Lec 10: Assumptions & Details
Syllabus:
(Click to download slides)
PART 1: Theory
Econometrics 1: Simple Linear Model
(1) Algebric Properties;
(2) Statistical Properties
Econometrics 2: Multiple Linear Model (Additional Materials)
(1) Algebric and Statistical Properties;
(2) Statistical Inference;
(3) Asyptotic Theory;
(4) Matrix Representation (Not required)
(1) Dummy Variable (As Independent & Dependent Variable);
(2) Function Form (Scale, Logarithm, Quadratic, Interaction; Misspecification Test);
(3) Heterogeneity in Parameters (Not required)
(4) Multiple Choice Model (Not required)
Econometrics 4: Heteroskedasticity
(1) Detect Hetroskedasticity (Residual Plot, Breusch-Pagan Test, White Test)
(2) Fix Heteroskedasticity (Robust Standard Error, Weighted Least Square, Feasible Generalized Least Square)
(1) Sources and Consequences (Selection, Omitted Variable Bias, Measurement Error, Simultaneous System)
(2) Instruemental Variable (Assumption & Estiamtion & Identification; Procedure; Applications)
(3) Matrix Form Representation (Not required)
(1) Stochastic Process
(2) Assumptions and Theorems (Standard and New)
(3) Details (Finite Distributed Lag Model, Time Trend and Seasonality, Serial Correlation)
(1) Fixed Effect v.s. Random Effect (FE estimator, FD estimatior, Demean, Quasi Demean)
(2) Difference in Difference (Assumptions, Validation, Triple Difference)
PART 2: Application
(1) Basics;
(2) Descriptive Statistics;
(3) Graphs
STATA 2: Data Generation Process
(1) FWL Theorem;
(2) Omitted Variable Bias;
(3) Measurement Error