ECONOMETRICS FOR PUBLIC POLICY ANALYSIS
2018/2019, Semester 2
Lee Kuan Yew School of Public Policy (Lee Kuan Yew School Of Public Policy)
Modular Credits: 4
The purpose of this course is to prepare students for becoming both critical consumers and competent producers of quantitative evidence used in the public policy arena. This course provides students with a solid grounding on economic theory and statistical techniques used to analyze public policy. At the end of the course, students will be able to use advanced econometric tools on real world policy problems and draw policy implications. The major topics covered include: inference and hypothesis testing, simple regression analysis, multiple regression analysis, non-linear regression models, binary dependent variable models, program evaluation, panel data analysis, and time series analysis and forecasting.
National University of Singapore
Lee Kuan Yew School of Public Policy
Econometrics for Public Policy Analysis
Second Semester AY18-19 Tues 2:00PM-5:00PM
Tan Poh Lin, Assistant Professor
Office hours: By appointment
This module is for students who have some background in statistics (PP5406 Quantitative Research Methods for Public Policy or equivalent) and want to acquire additional techniques and experience. Methods covered in this course include: multinomial and nested logit models, Heckman selection model, power analysis, regression discontinuity etc. Many classes will take the form of one hour of lecture followed by two hours of workshop, where groups of two or three students will be provided with a choice of datasets and independently test out the models and produce their own novel findings. At the end of the module, students will produce a research paper in response to a real-world policy paper or report, and identify potential clients who would value their findings.
The first objective is to strengthen students’ knowledge and familiar with additional statistical techniques:
multinomial and nested logit models, Heckman selection model, power analysis, regression discontinuity etc. Students will learn how to execute these techniques using STATA.
The second objective is to provide
students with practical experience in writing a research paper modelled after an actual policy paper or report. Students will gain training in presenting and defending their work in front of a skeptical audience, including the use of robustness and falsification tests.
PP5406 (Quantitative Research Methods for Public Policy) or equivalent
Readings will be provided. No textbook is required. You may use the following textbook (also used for PP5406) for students who would like to do additional reading:
Stock, James H. and Mark W. Watson.
Introduction to Econometrics
. Any edition.
You will need to bring a laptop with STATA installed for every meeting. We will hold STATA lab sessions or workshops in class every week. Please inform me if you need to borrow a laptop with STATA installed on it. If you want your own copy of Stata, plans for students at discounted prices are at
Grading and Requirements
The mid-term is on Tuesday 5 March 215pm-345pm. It will take place in lieu of class on Week 7. Laptops with STATA will be provided (no internet access). It is an open-book test. You may bring your lecture notes etc.
The assignment is due on Thursday 4 April by 5pm. You must submit it as a Word document on IVLE (
no PDFs please
). The folder will close at exactly 5pm.
It is maximum 3 pages.
More is not less.
Please use single spacing and font size 12 Times New Roman.
Presentations will take place on Tuesday 16 April 215pm-345pm, during class on Week 13. Print out a copy of your current draft for each group to review.
There are two deadlines:
Monday 11 March by 5pm. You must submit the names of your team (two or three), the policy paper or report that you will be addressing with your paper, identify at least one client that would be interested in your findings, and at least one potential dataset you plan to use, by email.
Monday 22 April by 5pm. You must submit it as a Word document on IVLE (
no PDFs please
). The folder will close at exactly 5pm.
It is maximum 15 pages (bibliography is required and not included in the page limit).
More is not less.
Please use single spacing and font size 12 Times New Roman. It must follow this format:
Background and findings of a real-world policy paper/report and the policy significance (suggest 1 page)
Your contribution through this paper (suggest 0.5 page)
Econometric model used to answer this question (suggest 1 page)
Dataset used for this paper, and summary statistics (suggest 1.5 pages)
Main results, including up to two tables (suggest 3 pages)
Robustness/falsification tests (suggest 1.5 pages)
Summary and conclusion (suggest 1 page)
Appendix: summary of the individual contributions of each student to the paper (one paragraph per student is sufficient). All students must contribute to data analysis.
It is important to take pride in your work. Imagine that your client will see this document, so try to be as professional and polished as possible.
On weeks where there is workshop, there will be a one-hour lecture. During workshop, students form groups of two or three.
At least one member of your group must be different each week.
Each group will be provided with a choice of datasets, and given 1.5 hours to choose one dataset (you may also provide your own), independently practice the techniques with the dataset, and produce your own novel findings. We will discuss our findings during the last 30 min of the workshop.
You must submit your Do file from each workshop.
Your grade for this component will be based on your three
No work will be accepted after the deadline. Exceptions due to personal or family emergencies will be considered on a case-by-case basis.
Academic dishonesty is taken very seriously at NUS and the LKY School. To avoid giving the impression that you are passing off other people’s work as your own, you will need to acknowledge conscientiously the sources of information, ideas, and arguments used in any of your assignments. In order to understand what counts as plagiarism and why it is wrong, students at the LKY School had taken the NUS online module on Academic Culture during the Orientation Programme and formally acknowledged that they had understood the contents. Students who would like an introduction to the different referencing styles can refer to the following website, among others: https://www.citethisforme.com/guides. You will be required to submit your research papers through IVLE for plagiarism checks.
Week 1 (15 Jan) – Module overview and STATA crash course
Week 2 (22 Jan) – Statistics
Week 3 (29 Jan) – Non-linearities I: log, logit, ologit, multinomial
Week 4 (5 Feb) – Non-linearities II: conditional and nested logit
Week 5 (12 Feb) – Panel data analysis
Week 6 (19 Feb) – Sampling weights and power analysis
Week 7 (5 Mar) – Mid-term exam
Week 8 (12 Mar) – IV and Heckman selection model
Week 9 (19 Mar) – Diff-in-diff
Week 10 (26 Mar) – Regression discontinuity
Week 11 (2 Apr) – Telling a story: checks and mechanisms
Week 12 (9 Apr) – Summary and reflections
Week 13 (16 Apr) – Class Presentations