課程名稱 |
應用統計與量化分析 Applied Statistics for Social Science Research |
開課學期 |
107-2 |
授課對象 |
社會科學院 公共事務研究所 |
授課教師 |
黃心怡 |
課號 |
GIPA5009 |
課程識別碼 |
343 U0090 |
班次 |
|
學分 |
2.0 |
全/半年 |
半年 |
必/選修 |
必修 |
上課時間 |
星期二3,4(10:20~12:10) |
上課地點 |
社科研605 |
備註 |
總人數上限:35人 |
Ceiba 課程網頁 |
http://ceiba.ntu.edu.tw/1072GIPA5009_ |
課程簡介影片 |
|
核心能力關聯 |
核心能力與課程規劃關聯圖 |
課程大綱
|
為確保您我的權利,請尊重智慧財產權及不得非法影印
|
課程概述 |
In this course, I will give a graduate level of introduction to regression models, which are commonly used in policy research, political science, and many other social science fields. This course focuses on linear regression models with interval variables (mostly cross-sectional analysis). The course starts with descriptive statistics. We will learn the basic properties of estimators, hypothesis testing, and the uses of many varieties of independent variables. We will also discuss the casual modeling as a way of understanding the inter-relationships among variables and of understanding why regression coefficients change. Then, the course will cover statistical inference, discussing the assumptions behind the classical linear regression models. Finally, we discuss the implications of a variety of violations of the classical assumptions. The course will also cover some basic concept of time-series model and limited variable models.
The course consists of lectures, discussions, exams, and homework assignments. We all know that the best policy to learn statistics is by doing it. Hence, there will be computer assignments to give you practical experience with using STATA and applying the statistical techniques.
|
課程目標 |
待補 |
課程要求 |
Class participation 10%
Homework 40%
Midterm examination 25%
Final examination 25%
作業需要使用STATA 12 or newer version |
預期每週課後學習時數 |
|
Office Hours |
|
指定閱讀 |
Wooldridge, J., Introductory Econometrics: A Modern Approach, 5th Ed, South-Western. (華泰代理) |
參考書目 |
待補 |
評量方式 (僅供參考) |
|
週次 |
日期 |
單元主題 |
第1週 |
2/19 |
Course Introduction
Introduction to probability
Wooldridge, Appendix B, Chapter 1
|
第2週 |
2/26 |
Bivariate regressions and multiple regressions: estimation and interpretation
Wooldridge, Chapter 2, 3, 7
Introduction to Stata I: basic syntax (資料匯入、清理、建立新變數)
|
第3週 |
3/05 |
Introduction to Stata II (continue) |
第4週 |
3/12 |
Interaction terms and curvilinear relationships
Wooldridge, Chapter 6, 7
|
第5週 |
3/19 |
Strength of Relationships and Inference in Regression Analysis
What is p-value?
Wooldridge, Chapter 4
課堂討論文章:
The fickle P value generates irreproducible results, Nature Methods, 2015
(https://www.nature.com/articles/nmeth.3288) |
第6週 |
3/26 |
Multiple Regression: Asymptotics and some other issues
Wooldridge, Chapter 5, 6, 7
課堂討論文章:
S.B. Dale and A.B. Krueger, “Estimating the Return to College Selectivity over the Career Using Administrative Earnings Data,” The Journal of Human Resources, 2014
Causal Modeling |
第7週 |
4/02 |
國定假日放假: 春假 |
第8週 |
4/09 |
Mis-specification, Measurement Error, and Missing Data
Wooldridge, Chapter 9 |
第9週 |
4/16 |
Midterm exam |
第10週 |
4/23 |
Instrumental Variable
Wooldridge, Chapter 15
課堂討論文章:
|
第11週 |
4/30 |
Heteroskedasticity & Multicollinearity
Wooldridge, Chapter 8 and 3.4
|
第12週 |
5/07 |
Time-Series Analysis
Wooldridge, Chapter 10
|
第13週 |
5/14 |
Panel data analysis
Wooldridge, Chapter 13-14
Ashenfelter, O., and A. Krueger. "Estimates of the Economic Returns to Schooling from a New Sample of Twins." American Economic Review 84, no. 5 (December 1994): 1157-1174.
|
第14週 |
5/21 |
Catch up and perhaps some treatment effect estimation |
第15週 |
5/28 |
研究週 |
第16週 |
6/04 |
Limited dependent variable models
Wooldridge, Chapter 17 |
第17週 |
6/11 |
Catch up or other issues in logit/probit models |
第18週 |
6/18 |
Final exam |
|