Mini-course in Machine Learning and Causal Inference

台大經濟系 微課程 – 機器學習與因果推論

台大經濟系 111-1學期「微課程- 機器學習與因果推論」 課程資訊如下：

上課時間：2022年 8月29日 (一) 至 9月2日 (五) 密集授課，18小時。

8/29 – 9/1：上午 9:00 – 11:00 & 下午 14:00 – 16:00

9/2：上午 9:00 – 11:00

授課教師：陳釗而

中文課名：微課程 – 機器學習與因果推論

英文課名：Mini-course in Machine Learning and Causal Inference (Department of Economics, National Taiwan University)

課號：ECON5188。 課程識別碼：323 U1090 (授課主要對象為大學部三年級四年級學生)

學分數：1

選課人數上限：30人

地點：遠距授課

台大經濟系 110-2學期「微課程- 機器學習與因果推論」課程資訊如下：

上課時間：2022年 2月8日 (二) 至 2月11日 (五)，每日上午 9:00 – 11:00 & 下午 14:00 – 16:00 密集授課。

中文課名：微課程 – 機器學習與因果推論

英文課名：Mini-course in Machine Learning and Causal Inference

課號：ECON5188。 課程識別碼：323 U1090 (授課主要對象為大學部三年級四年級學生)

選課人數上限：30人

地點：遠距授課

Course Description

The course discusses machine learning as well as the use of these methods for causal inference in economics. The challenging part of empirical research in a data-rich environment is to raise good questions and do good data. To this end, we go through examples of the off-the-shelf applications of machine learning to economics. We then present highlights and empirical studies from the emerging econometric literature combining machine learning and causal inference. Mastery of techniques taught in classes demonstrated through the completion of 5 assignments.

Students should do problem sets and assigned readings. We encourage questions and class discussion – we will be asking you questions too!

Course Objective

You will finish the course equipped with a workman’s familiarity with the causal machine learning techniques, facility with data handling, and programming.

Course outline

Topic 1: A helicopter tour of machine learning in economics

Topic 2: Regression

Topic 3: Research designs and empirical strategies: The Furious Five

Topic 4: Logistic regression

Topic 5: Modern high-dimensional econometrics

Topic 6: Double lasso selection procedure

Topic 7: Decision trees

Topic 8: Random forests

Topic 9: Approximate factor models

Topic 10: Factor models in causal inference

Topic 11: Double machine learning procedure

Topic 12: Causal forests

Topic 13: Heterogeneous treatment effects and policy learning

Topic 14: Quantile treatment effects

Topic 15: Machine learning methods that economists should know about

Topic 16: Discussions

Required readings

Textbook:

1. James, Witten, Hastie, and Tibshirani (2021). An Introduction to Statistical Learning with Applications in R. 2^nd ed. Springer. [Topics 4, 5, 7, and 8]
2. Taddy (2019). Business Data Science: Combining Machine Learning and Economics to Optimize, Automate, and Accelerate Business Decisions. McGraw-Hill. [Topics 2, 3, 4, 11, and 12]
3. Angrist and Pischke (2009). Mostly Harmless Econometrics. Princeton University Press. [Topics 2, 3 and 14]

The instructor’s lecture slides/notes. [Topics 1-14]

Papers:

1. Athey, S. and G.W. Imbens (2019). “Machine learning methods that economists should know about,” Annual Review of Economics, 11, 685-725. [Topic 15]
2. Shah, V., N. Kreif and A.N. Jones (2021). “Machine learning for causal inference: estimating heterogeneous treatment effect,” Chapter 16, Handbook of Research methods and applications in empirical microeconomics. Edward Elgar Publishing. [Topic 13]

Grading

Students will do 5 assignments (100%). In addition, students will be asked to demonstrate their analytical and computer-based solutions to their fellow students.