短期课程1:High-Dimensional Econometrics Data Analysis
Course de1111ion:

High-dimensional data analysis techniques are introduced for studying economic and financial data. Topics on dimension reduction, including feature selection and feature extraction are covered. Statistical software is used during the class and real data examples are discussed to illustrate the application of high dimensional statistical methods in practice.
Schedule (Tentative)
Day 1: Introduction(10月13日,上课地点:明新0407)

Class 1: Examples of high-dimensional econometrics data

Class 2: Prediction problems and classification problems

Class 3: Impacts of high-dimensional data to statistical methods

Class 4: Ideas of dimension reduction
Day 2: Feature Extraction Approach of Reduction(10月19日,上课地点:上午明新0407,下午明法0202)

Class 1: Principal component analysis and linear discriminant analysis

Class 2: Factor models

Class 3: Applications in capital asset pricing models

Class 4: Applications in arbitrage pricing theory
Day 3: Feature Selection Approach of Reduction(10月20日,上课地点:明xin 0407)

Class 1: Bayesian approach versus ridge/LASSO/SCAD regression

Class 2: Model selection via information criteria (AIC/BIC)

Class 3: Numerical methods for non-smooth function optimization

Class 4: Feature selection methods of identifying causality in econometric data
Day 4: Introduction to Artificial Intelligence Approach(10月26日,上课地点:明主0305)

Class 1: Logistic regression

Class 2: Artificial neural network

Class 3: Support vector machine

Class 4: Deep learning

If time permits, the topic of management of large scale portfolio will be also covered: (1) Traditional Markowitz theory revisited; (2) Recent developments of high-dimensional methods; (3) Portfolio selection by combining LASSO and Markowitz theory; (4) Special techniques for high-dimensional sample covariance matrix.

吴自添(Chi Tim, Ng), 韩国全南大学副教授。Tim博士于2007年毕业于香港中文大学统计系,师从国际知名统计学家 Ngai Hang, Chan教授。他的主要研究方向为高维数据分析,时间序列分析,变量选择,随机积分以及期权定价等。目前,他已经在国际知名统计学期刊上发表30多篇论文。

短期课程2:Econometrics of Program Evaluation
Course de1111ion

The main purpose of this course is to provide students with the knowledge of program evaluation techniques with focus on economic applications. The field of program evaluation may be called causal inference. The field of program evaluation consists of two closely related parts: identification and actual implementation. The identification part of program evaluation concerns how and under what assumptions we can identify the causal effect of treatments. It uses specialized concepts and also has a direct connection with research design. The actual implementation part of program evaluation is more closely related to typical econometrics and it concerns estimation and statistical inference. Program evaluation is arguably the current center of applied research in economics. Any economist needs to acquire familiarity with the identification part. Students who want to do applied researches are also expected to be familiar with the actual implementation part.
Schedule (Tentative)
Day 1

Class 1: Course outline and basic concepts

Class 2: Randomized controlled trials

Class 3: Multiple testing problems

Class 4: Estimation with unconfoundedness assumption

Class 5: Inverse probability weighting estimation
Day 2

Class 1: Doubly robust estimation

Class 2: Matching

Class 3: Conditional average treatment effects

Class 4: Instrumental variables estimation, one-sided non-compliance

Class 5: Instrumental variables estimation, two-sided non-compliance
Day 3

Class 1: Marginal treatment effects

Class 2: Marginal treatment effects with discrete instruments

Class 3: Regression discontinuity design, identification

Class 4: Regression discontinuity design, implementation

Class 5: Fuzzy regression discontinuity design
Day 4

Class 1: Difference-in-differences

Class 2: More topics on difference-in-difference

Class 3: Oxaca-Blinder decomposition

Class 4: Bounding treatment effects

Class 5: Empirical welfare maximization

Ryo Okui, Associate Professor, Seoul National University. Okui is a well-known professor in econometrics. His research interests include Microeconometrics, Applied Microeconomics, and Experimental Economics. His papers have been published in many top economic journals such as Econometrica, Review of Economic Studies, and Journal of Econometrics. He has received many awards and honors such as in September 2010, he was awarded the JSS Ogawa Award by Japan Statistical Society, and in April 2005, he was awarded The Hiram C. Haney Fellowship Award in Economics by University of Pennsylvania. For more information about him, please visit his personal website: https://sites.google.com/site/okuiryoeconomics/.


人大经济论坛 rdjjlt.org 人大经济论坛 bbs.econ.ruc.edu.cn