**References:**

Jeffery Wooldridge: Introductory Econometrics: A Modern Approach, second edition, Tsinghua University Press, 2004

**Structure of the Course:**

1. The Nature of Econometrics and Economic Data

2. Review of Probability and Statistics

3. The simple Regression Model

3.1 Definition of the Simple Regression Model

3.2 Motivation for Multiple Regression

3.3 Mechanics and Interpretation of Ordinary Least Squares

3.4 The Expected Values of the OLS Estimators

3.5 The Variance of the OLS Estimators

3.6 Efficiency of OLS: The Gauss-Markov Theorem

4. The Multiple Regression

4.1 Sampling Distributions of the OLS Estimators

4.2 Testing Hypothesis About a Single Population Parameter: The *t* test

4.3 Confidence Intervals

4.4 Testing Hypotheses About a Single Linear Combination of the Parameters

4.5 Testing Multiple Linear Restrictions: The *F* Test

4.6 Reporting Regression Results

5. Large Sample Problems

5.1 Consistency

5.2 Asymptotic Normality and Large Sample Inference

5.3 Asymptotic Efficiency of OLS

6. Model Specification and Data Problem

6.1 Functional Form misspecification

6.2 Using Proxy variables for unobserved explanatory variables

6.3 Properties of the OLS Under Measurement Error

6.4 Missing Data, Nonrandom Samples, and outliers

7. Dummy Variable

7.1 Describing Qualitative Information

7.2 A Single Dummy Independent Variable

7.3 Using Dummy Variable for Multiple Categories

7.4 Interactions Involving Dummy Variables

7.5 A Binary Dependent Variable: The Linear Probability Model

8. Time Series Regression

8.1 Properties of OLS with Serially Correlated errors

8.2 Testing for Serial Correlation

8.3 Correcting for Serial Correlation with Strictly Exogenous Regressors

8.4 Differencing and Serial Correlation

8.5 Heteroskedasticity in Time Series Regression

8.6 Serial correlation robust inference after OLS

**Course Objectives:**

This course will, on the one hand, introduce basic theories and methods of Econometrics, on the other hand, emphasize on how these methods can be used to solve practical problems in the empirical study. For this purpose, the course will focus on the introduction of multiple lineal regressions in terms of cross-section data. In addition, the emphasis includes how to design and test econometric models. By learning this course, students are required not only to master fundamental theories of Econometrics, but also to cultivate the handling capability. At the end of the course, students should be capable of using STATA to make empirical analysis on some interesting economic phenomenon.

**Examination:**

Regular attendance, participation, 2 written homework: 10%

2 computer assignments (STATA): 10%

1 Group Essay: 15% (Requirement: Select one research question; Design Questionnaire; Network sampling; Data collection; Specific econometric model; Estimation; Testing model)

Mid –Term Test: 30%

Final Test: 35%

**Credits & Workload:**

4 Credits & 4 hours per Week (teaching) + 2 hours per Week (tutoring); 14 Weeks, 56 hours (teaching) + 28 hours (tutoring)

**Excerpt:**

Notes of Simple Regression

1 Motivations of regression

The relevance among variables of interest (The present scope is confined to the relationship between Y and X in first chapter, which can be readily extended to multi‐variable case in sequential study.):

• What is (if there exists) the relationship between wage and educational level?

• Does reducing class size improve elementary school education?

• Is there racial discrimination in the labor market?

• What will economic growth of China be next year/next decade?

2 Fundamentals of regression

Quantitative questions, quantitative answers

Data: Sources and Types

(1) Experimental versus Observational Data

(2) Cross‐sectional/Time series/Panel Data

*There exists another kind of panel (longitudinal) data: pseudo panel.

3 Issues of OLS

It is imperative for us to be quite familiar with SLR.1‐SLR.5 as well as their disparities.

It is worthwhile to distinguish respective conditions under which different conclusions hold.

E.g.:

The property of unbiasedness has nothing to do with SLR.5;

The following question arises naturally: what role does SLR.5 play in our story?

Kinds of definitions should also be paid certain attention, such as PRF, SRF, fitted/predicted values, and disturbance versus residual.

It is desirable to grasp the basic algebraic operations, such the deduction of series of properties OLS estimator.