Embedded Files
Fall 2015
Goal: The course will provide an introduction to the statistical tools designed to evaluate causal effects of public programs.
By the end of the course, students should be able to conduct and to critically assess the validity of program evaluation studies.
This is a methodological course.
The foundational basis for this course is BUEC333. You must have taken BUEC333 to be able to take this course.
A precise mathematical understanding of concepts is crucial to good applications.
Topics we will cover include: randomized experiments, observational studies with and without ignorable treatment assignment, instrumental variables, regression discontinuity, and sensitivity analysis.
Survey articles:
Angrist and Krueger (1999) Empirical Strategies in Labor Economics.
Blundell and Costa-Diaz (2009) Alternative Approaches to Evaluation in Empirical Microeconomics.
Imbens and Wooldridge (2009) Recent Developments in theEconometrics of Program Evaluation.
Information on R:
R can be downloaded for free from its official website. RStudio (interface for R) can be downloaded here.
To learn R in R, check out swirl. Entertaining intro to R here. Less entertaining intro to R here. R by example.
Free book via SFU library. Check out the Intro, Basics, and Linear Regression.
Take-home exams:
Schedule to Exam 1 - in class exam on Sept 29
Lecture 1 09/08. Causality. The potential outcome framework for causal inference. The Rubin Causal Model.
Reading
(1) Up to 1.6 in Ch 1 in Imbens and Rubin book.
Extra readings:
(1) Holland (1986) pdf, (2) Little and Rubin (2000) pdf, (3) Rubin (2005) pdf, (4) Winship and Morgan (1999) pdf
Lecture 2 09/15. Average treatment effects with and without randomization. Unconfoundedness and olverlap. Selection bias.
Readings
(1) Chapter 1 in Mastering Metrics
(2) Chapter 2 in Mostly Harmless Econometrics (MHE)
(3) Up to Section 2.3.2 in Duflo et al. (2008)
(4) Selection bias blog
Extra readings:
(1) Section 3.2 in Duflo and Kremer
(2) Up to Section IV in Blundell and Costa-Diaz (2009)
HW Homework due 9/22
link
solution
(1) Returns to Plastic Surgery in Marriage and Labor Markets by Lee and Ryu
(2) Beauty and the Labor Market by Hamermesh and Biddle
Lecture 3 09/22. More on unconfoundedness and overlap. Selection bias. ATE and linear regression.
Extra reading:
Exam 1 09/29. Up and to including Lecture 3.
Exam and solutions
Max 18, Min 5, Average 15.31 (77%), Std dev 3.66 (18%)
Schedule to Exam 2 - take home exam due Nov 10
Lecture 4 10/06. Causal effects and regression analysis.
Reading
(1) Section 2.2.1 in Angrist and Krueger (1999)
Extra reading:
(1) Sections 9.2, 9.5, and 9.6 here
Lecture 5 10/13. More on regression analysis. Matching. Causal graphs. IV.
Reading
(1) Sections 3.2, 3.3.1, 4.1 (up to 4.1.2) in Mostly Harmless Econometrics
(2) Chapter 3 in MM
(4) Section 2.2.3 in Angrist and Krueger (1999)
Lecture 6 10/20. Instrumental variables.
Readings
(1) Using IVs to learn more from social policy experiments up to page 11
Extra reading:
(1) Section 2.5.4 here
HW Homework due 10/27
Schedule to Exam 3 - take home exam due Nov 24
Lecture 8 11/03. Difference-in-differences.
Reading
(1) Section 2.2.2 in Angrist and Krueger (1999)
(2) Section 5.2 in Mostly Harmless
(3) Youtube videos
Panel data, parallel trends, and DID 1 2 3
Applications:
Effect of Minimum Wage on Employment video
Effect of Immigration on Employment video
Lecture 9 11/10. Exam 2 due.
Lecture 10 11/17. More on DID.
Lecture 11 11/24 Exam 3 due. Fixed effects.
Korenman and Neumark (1991): Does Marriage Really Make Men More Productive?
Presentations 12/01. Papers for presentations:
Children and Their Parents' Labor Supply: Evidence from Exogenous Variation in Family Size
Do Better Schools Matter? Parental Valuation of Elementary Education.
Minimum Wages and Employment: A Case Study of the Fast Food Industry in New Jersey and Pennsylvania
Estimating the Payoff to Attending a More Selective College
The Effect of Prison Population Size on Crime Rates: Evidence from Prison Overcrowding Litigation
Worker’s Compensation and Injury Duration: Evidence from a Natural Experiment
How Large Are the Effects from Changes in Family Environment? A Study of Korean American Adoptees
Aaronson (1998): Using Sibling Data to Estimate the Impact of Neighborhoods on Children's Educational Outcomes
Bogart and Cromwell (1999): How Much Is a Neighborhood School Worth?
Black et al. (2003): Is the Threat of Reemployment Services More Effective Than the Services Themselves? Evidence from Random Assignment in the UI System.
Fikelstein et al (2012): The Oregon Health Insurance Experiment: Evidence from the First Year
Google Sites
Report abuse