中级计量经济学 Intermediate Econometrics
Methods in identifying (causal) relationships in social science
- ❗️Announcement❗️
- Syllabus
- Materials
- Lecture 0. Introduction
- Lecture 1. Finite Sample OLS
- Lecture 2. Large Sample Theory
- Lecture 3. Heteroskedasticity, Autocorrelation and Generalized Least Squares
- Lecture 4. Maximum Likelihood Estimation
- Lecture 5. Specification Error, Measurement Error, and other Data Problems
- Lecture 6. Instrumental Variable and Generalized Methods of Moments
- Lecture 7. Simultaneous Equations Model, Panel Regressions
- Lecture 8. Dynamic Panel Regressions
- Lecture 9. Monte Carlo Methods and Bootstrap
- Lecture 10. Random Experiments and Natural Experiments, Treatment Effects
- Lecture 11. Time Series Models
- Lecture 12. Machine Learning and Econometrics
❗️Announcement❗️
2021.12.10: 2020考试题
2021.12.03: 作业4. 数据. 参考答案：RDD, Housing
2021.11.12: 请确认复制论文选题后在此列表中登记选题：复制论文列表.
Syllabus
This course is an applied econometrics course, in the sense that mathematical details are less emphasized. That being said, students are expected to know undergraduate level linear algebra, calculus, probability and statistics. The main contents are:
- Finite Sample OLS
- Large Sample Theory
- Maximum Likelihood Estimation
- Endogeneity Problems in Economics
- Instrumental Variables and Generalized Methods of Moments
- Panel Regressions
- Random Experiments and Natural Experiments
- Treatment Effects
If time permits, we’ll also briefly discuss the following topics: simultaneous equations models, binary-response models, machine learning fundamentals. The goal is to let students know both the big picture and necessary details of applying statistical (including econometrics and machine learning) models to problems in social sciences, especially economics and finance.
这门课是 应用 计量经济学。因此，除了关键步骤，数学推导将尽量减少。但微积分、线性代数、概率论和统计的主要知识仍然必不可少。主要内容如下：
- 小样本OLS及其性质
- 大样本OLS
- 最大似然估计
- 内生性问题的几种来源
- 工具变量，GMM
- 面板回归
- 蒙特卡洛模拟和自助法
- 随机实验和自然实验
- 处理效应
如果时间允许，我们也会简要讨论以下一些内容：联立方程模型，二值选择模型，机器学习基础。讨论这些内容的原因：让学生对社会科学研究中的问题的性质，以及如何根据这些性质选择合适的数量模型，有比较明确的认识。
Text Book and Useful References
Text book: 陈强, 《高级计量经济学及Stata应用》，第2版，http://www.econometrics-stata.com/col.jsp?id=101
References:
- Hayashi, F. (2000). Econometrics. Princeton University Press.
- Wooldridge, J. M. (2010). Econometric Analysis of Cross Section and Panel Data, 2nd ed.
- Angrist, J. D., and J. Pischke, 2009, Mostly Harmless Econometrics: An Empiricist’s Companion. 1 edition. (Princeton University Press).
Grading
- 3 to 4 exercises, 20%
- Replication project, 40%. Requirement
- Final exam, 40%
PLAGIARISM IS STRICTLY PROHIBITED. You may immediately fail the course if copy-pasting other’s work. Discussion is, of course, permitted.
抄袭作业零容忍。抄袭他人作业可能会直接挂科。 讨论、交流没有问题，但仍需自己完成。
Materials
Lecture 0. Introduction
Lecture 1. Finite Sample OLS
Lecture 2. Large Sample Theory
Lecture 3. Heteroskedasticity, Autocorrelation and Generalized Least Squares
Lecture 4. Maximum Likelihood Estimation
hand-written notes of Lec 4 and 5, Oct.15
Lecture 5. Specification Error, Measurement Error, and other Data Problems
annotated slides, Oct.22 updated
Lecture 6. Instrumental Variable and Generalized Methods of Moments
annotated slides, Nov.6 updated
Efficiency of 2SLS vs. IV when homoskedasticity
Lecture 7. Simultaneous Equations Model, Panel Regressions
Annotated Panel Slides, Nov.12 updated
Lecture 8. Dynamic Panel Regressions
Lecture 9. Monte Carlo Methods and Bootstrap
annotated slides, Monte Carlo and Bootstrap, Nov.19
Lecture 10. Random Experiments and Natural Experiments, Treatment Effects
1. annotated Experiments Slides, Nov.19
2. Binary Choice Models Slides
2. annotated Binary Choice Models Slides, Nov.19
annotated 3. Treatment Effects Slides
Lecture 11. Time Series Models
1. Annotated Stationary process
2. Unit root and cointegration
2. annotated unit root and cointegration
Lecture 12. Machine Learning and Econometrics
The book I mentioned in class for learning ML (strongly recommend): https://www.oreilly.com/library/view/hands-on-machine-learning/9781492032632/. The author provides complete hands-on code here: https://github.com/ageron/handson-ml2
Classical references on the statistical models: An Introduction to Statistical Learning, The Elements of Statistical Learning. 中文参考书：西瓜书