中级计量经济学2班——2012年秋季学期双学位课程介绍及教师简介

发布日期:2012-09-04 12:55    来源:北京大学国家发展研究院

任课教师(Lecturer): 张丹丹

先修课要求(Prerequisite):微积分、线性代数、概率统计

 

课程简介(Course Description):

计量经济学探讨的是用统计技术研究经济数据的方法,目的是通过对微观数据的分析寻找经济问题的因果解答。本课程的教学内容包括经典OLS线性回归模型、多元线性回归模型、工具变量方法、和面板数据分析方法等。本课程强调利用统计软件(STATA)实际操作微观数据分析现实经济问题。通过这门课的学习,使学生掌握如何进行实证研究,以及如何评断实证研究中存在的问题。

Econometrics is the statistical analysis of economic (and related) data. This course is about using data to measure a causal effect.  Topics studied will include classical and multiple regression methods, instrumental variables regression and regression with panel data. The course emphasizes economic applications and hands on data analysis using STATA. The objective of the course is for the student to learn how to conduct – and how to critique – empirical studies in economics and related fields.

 

教学目的 (Learning Outcomes):

通过这门课的学习,使学生能解释计量经济学技术的原理,利用数学公式表达计量方法,阐述计量方法存在的问题,并能用计量软件STATA分析微观数据,解释结果并进行统计检验。

Upon successful completion of the requirements for this course, students will be able to:

explain the intuition of the econometric techniques discussed, formulate the econometric techniques in mathematical terms, describe the properties and problems of the econometric techniques, use STATA to study actual data sets, interpret empirical results and perform tests.

 

教科书 (Textbook):

Introductory Econometrics: A Modern Approach, Jeffery M. Wooldridge, Fourth Edition, 2009.

[Jeffery M. Wooldridge, 计量经济学导论:现代观点(第4版)(清华经济学系列英文版教材), 清华大学出版社,2009年7月]

 

评分标准 (Grading):课堂参与(Course participation):5%;作业(Problem sets):25%;期中考试(Midterm exam):30%; 期末考试(Final exam):40%

 

教学大纲(Course Outline):

 

1. 导论 (Overview)

(1) 教学内容介绍(Brief overview of the course)

(2) 统计概率基础回顾 (Review of probability and statistics)

 

2. 简单线性回归模型 (Simple Regression Model)

  (1) 基本概念 (The basic concept)

(2) 普通最小二乘法(OLS)估计值的推导 (Deriving the OLS estimates)

(3) OLS的性质 (Properties of OLS statistics)

(3) 双变量线性回归模型的假设 (Assumptions of the bivariate linear regression model)

(4) OLS估计值的期望和方差 (Expected values and variances of the OLS estimators)

 

3. 多元回归模型的估计 (Multiple Regression Analysis: Estimation)

(1) 基本概念 (The basic concept)

(2) 获得OLS估计值 (Obtaining the OLS estimates)

(3) OLS的性质 (Properties of OLS statistics)

(4) 多变量线性回归模型的假设 (Assumptions of the multivariate linear regression model)

(5) OLS估计值的方差 (The variance of the OLS estimators)

(6) 估计误差项的方差 (Estimating the error variance)

(7) 无偏性和遗失变量偏差 (Unbiasedness and omitted variable bias)

(8) 高斯-马尔科夫定理 (The Gauss-Markov Theorem)

 

4. 多元回归模型的推断 (Multiple Regression Analysis: Inference)

(1) OLS估计值的样本分布 (Sampling distributions of the OLS estimators)

(2) 单个总体参数和参数线性组合的假设检验:t检验 (Testing hypothesis about a single population parameter or a single linear combination of the parameters: The t test)

(2) 多个线性约束的假设检验:F检验 (Testing multiple linear restrictions:The F test)

 

5. 多元回归进一步讨论 (Further Issues)

  (1) 数据单位变化对OLS估计值的影响 (Effects of data scaling on OLS Statistics)

  (2) 标准化系数 (Standardized coefficients)

  (3) 方程形式 (Functional Forms)

  (4) 含交叉项的模型(Models with interaction terms)

 

6. 多元回归虚拟变量 (Multiple Regression Analysis with Qualitative Information)

(1) 解释变量中有一个虚拟变量 (A single dummy independent variables)

 (2) 虚拟变量和对数被解释变量(Dummy variables and logged dependent variables)

 (3) 多重虚拟变量(Using dummy variables for multiple categories)

  (4) 虚拟变量的交互项 (Interactions involving dummy variables)

  (5) 被解释变量是虚拟变量: 线性概率模型 (The linear probability model)

 

7. 异方差 (Heteroskedasticity)

(1) 估计稳健标准误 (Estimating robust standard errors)

(2) 异方差的检测 (Testing for heteroskedasticity)

(3) 加权最小二乘法估计 (Weighted Least Squares estimation)

(4) 可行广义最小二乘法(Feasible Generalized Least Squares) 

 

8. 模型设定和数据问题 (More on Specification and data issues)

  (1) 方程形式误设(Functional Form Misspecification)

  (2) 指代变量(Proxy variables)

  (3) 测量误差(Measurement Error)

  (4) 非随机样本和离群点 (Nonrandom samples and outliers)

 

9. 工具变量估计 (Instrumental variables estimation)

(1) 有效工具变量需满足的两个条件 (Two conditions for a valid IV)

(2) 两阶段最小二乘法 (Two Stage Least Square, TSLS)

(3) TSLS估计值的一致性 (Consistency of the TSLS estimator)

(4) 工具变量有效性的检验 (Checking instrument validity)

 

10. 面板数据回归技术 (Panel data)

(1) 利用面板数据进行回归分析 (Regression analysis with panel data)

(2) 固定效应回归技术 (Fixed effects regressions: within-groups)

(3) 差别差分(Differencing in differencing)

(4) 随机效应回归技术(Random effects regressions)

 

教师简介见附件 pdf CV_DZhang_Aug2012.pd f