ECON 8848-001 Applied Microeconometrics

by user

Category: Documents





ECON 8848-001 Applied Microeconometrics
University of Colorado at Boulder
Department of Economics
Prof. Brian Cadena
[email protected]
(303) 492-7908
Website: Desire2Learn
ECON 8848: Applied Microeconometrics, Spring 2013
Syllabus and Schedule
Office Hours: MW 1:30-3:30 PM
Economics 14A
Other times by appointment
Course Description:
Students who are successful in this course will be well prepared to conduct empirical research
across a broad range of fields, although the tools are used most frequently in the applied
microeconomics fields. The course provides a “user’s guide” to many of the most commonly
used econometric techniques, with a heavy focus on implementation and interpretation. We will
begin the course with a STATA boot camp, quickly becoming familiar with the software
package including programming techniques and data management skills. We will then move
through a range of econometric topics, making sure to practice each technique in STATA. I
hope to live up to the following quotation by Edward Leamer in his article Let’s Take the Con
out of Econometrics (AER, 1983):
“Methodology, like sex, is better demonstrated than discussed, though often better
anticipated than experienced.”
To enroll in this course, you must have a working knowledge of statistics and econometrics
equivalent to that obtained in ECON 7818 and ECON 7828.
Course Materials:
There is no required textbook for this course, although I will provide references to a number of
books and articles for the interested student. We will also read and discuss several articles.
Some of these articles will be “theory” articles, discussing the relative merits of estimators or
developing and applying new ones. Others will be “application” papers, usually papers that use
a technique we have discussed in an honest and useful way. I will also provide lecture notes,
and you will find these and the assigned articles posted or linked on the Desire2Learn website.
You should read the articles assigned prior to coming to class and be prepared to answer
questions and participate in discussions. Please bring a copy of the papers we are discussing
with you to class.
Students are not required to purchase their own copies of STATA, although those desiring to do
so qualify for a substantial discount through the University’s GradPlan. More information is
available through a link posted on the Desire2Learn website. I recommend Stata/IC. The price is
$179 for a perpetual license (one that never expires).
You will receive a copy of the STATA documentation in PDF format if you choose to purchase
your own. If you expect to use STATA beyond this course, you can feel free to purchase a more
advanced copy (SE or MP), but the Intercooled version will allow you to complete all the
requirements of this course.
I will use STATA during some lectures to demonstrate estimators and methods that we cover. If
you have STATA installed on a laptop, you may find it useful to bring on those days.
Requirements and Grading:
Your grade will depend on your performance on a number of assignments, according to the table
Problem Sets
Paper Replication/Extension
Final Exam
Due Date
Every 1-2 weeks (~10 total)
Friday, 5/3 5 PM Hard Copy in my Office
Wednesday, 3/13 In Class
Thursday, 5/9 7:30-10:00 AM
Problem Sets will be assigned roughly weekly, and will be closely related to the topics covered
in this course. These problem sets will allow you to gain direct experience with all of the
econometric techniques we cover. All assignments will be STATA-based, although they will
require answering interpretation/“thinking” questions as well. Although I may ask you to prove
something formally, this is much less likely than asking you to simulate or to demonstrate a
particular property using real-world or simulated data. Problem Sets will be graded on a 3-point
scale, with missing assignments given a grade of zero.
Paper Replication/Extension: Unlike the harder sciences, the field of economics places a
relatively small weight on the value of replication. Nevertheless, economists make mistakes all
the time, and some of them go undiscovered forever. So, as a means to practice all of the skills
we are developing, and in service of the broader good, you will replicate a paper in a field that is
of interest to you. You should choose a published paper that relies on publicly available data or
on data that the authors have made freely available. You should also provide at least one
extension to the original work. Possible extensions include adding additional years of data,
running additional specifications (e.g. functional form, RD instead of DiD, etc.), and subjecting
the results to additional robustness checks. Alternatively, you could use similar methods in a
slightly different context – different geography, different time period, etc. A hard copy of this
paper will be due in my office by close of business on May 3.
The Midterm will cover material from the beginning of the course through lecture on March 11.
The exam will take place during our normal class meeting on Wednesday, March 13. You will
not have to do any STATA programming for the midterm. Instead, the questions will focus on
the interpretation and implementation of techniques we have discussed. I will ask several
questions of the form: “Suppose you used to technique X to answer question Y and received the
following results. What do they mean? What are you worried about? How could you address
potential threat to identification Z? You may also be asked questions about the papers that we
The Final Exam will be nominally cumulative, but it will focus heavily on material covered
after the midterm. It will be similar in format to the midterm. Our assigned time from the
Registrar is 7:30-10:00 AM on Thursday, May 9. University policy provides students with three
or more exams on the same day the right to reschedule exams following the first two.1 Any
student wishing to invoke this right should notify me as soon as possible and no later than
February 28. I will ask for a printed copy of your schedule to verify the conflict.
Final Letter Grades will be a weighted average of each of the components listed above. Prior
to averaging, I will assign letter grades to each component based on the scores a good student at
this level could reasonably be expected to attain.
Seminar Series: You are strongly encouraged to attend the Economics department seminar
series, especially when the speaker presents on an empirical applied micro topic. Learning to
conduct and present original research is the key to your success in the discipline. These seminars
are an excellent resource for you in that endeavor.
Late Assignments/ Missed Examinations Policy: Problem Sets will be turned in through the
Desire2Learn website where they will receive a time stamp. You will have no less than one
week to work on a given problem set, and each will be due on a Friday by 5 PM. Following a 5minute grace period, I will deduct 1 point from each assignment for each day it is late.
Assignments submitted more than 48 hours after the due date will receive no credit.
The paper replication must be turned in on time. I will deduct 25 percent of the grade for each
day after the deadline when it is turned in.
If you miss the midterm or the final exam you will receive no credit unless you provide
documentation of a medical or family emergency. In the case of a legitimate emergency, the
missed exam will be given no weight in the calculation of the final grade and other assignments
will be reweighted accordingly. There will be no make-up exams. If you foresee any conflict
that will prevent you from taking an exam, please let me know as soon as possible and at least
two weeks beforehand.
A note on my role: I will always be willing to offer you assistance with any assignment for this
course, including the final paper. I will strongly suggest, however, that you form study groups
for the problem sets and use the other members of your group as your initial resource in solving
programming problems. In addition, I cannot generally offer help on projects that are unrelated
to this course, e.g. work you are doing as part of your dissertation or as an RA for other faculty
members. My goal in offering this course is to create a critical mass of well-trained graduate
students who can then continue to learn more on their own and begin to serve as a resource to
each other.
Tentative Schedule
Tentative Dates
Introduction and STATA Basics
1/14, 1/16
Advanced STATA
Descriptive Statistics, Figures and Tables
Programming – Loops, Macros
Linear Regression Review
Functional Forms – Logs, Polynomials,
Categorical Variables, Interaction Models
Review of FWL and the meaning of
“controlling for”
The Experimental Ideal
Treatment Effects – Potential Outcomes
Causality in an OLS Regression – the CIA
Propensity Score Methods
Advanced Data Management
2/4, 2/6
Panel Data Models
2/27, 3/4 – paper
3/6, 3/11 – paper(s)
NO CLASS – Individual Meetings
3/16, 3/18
NO CLASS – Spring Break
3/23, 3/25
Variance Estimation in Panel Models
4/1, 4/3, 4/8 - papers
Instrumental Variables
Basics – Constant Treatment Effects
4/10, 4/15 - paper
Local Average Treatment Effects
Regression Discontinuity
4/22, 4/24 – papers
Binary Dependent Variables
5/4 4:30-7:00 PM
Other University Policies:
Disability Accommodation
If you qualify for accommodations because of a disability, please submit to your professor a
letter from Disability Services in a timely manner (for exam accommodations provide your letter
at least one week prior to the exam) so that your needs can be addressed. Disability Services
determines accommodations based on documented disabilities. Contact Disability Services at
303-492-8671 or by e-mail at [email protected]
If you have a temporary medical condition or injury, see Temporary Injuries under Quick Links
at Disability Services website (http://disabilityservices.colorado.edu/) and discuss your needs
with your professor.
Religious Observances
Campus policy regarding religious observances requires that faculty make every effort to deal
reasonably and fairly with all students who, because of religious obligations, have conflicts with
scheduled exams, assignments or required attendance. In this class, please inform me no later
than two weeks prior to any conflict you foresee, and sooner if possible, so that we can find an
alternative arrangement for you to complete the requirements of the course. See full details at
Classroom Behavior and Gender Identity
Students and faculty each have responsibility for maintaining an appropriate learning
environment. Those who fail to adhere to such behavioral standards may be subject to discipline.
Professional courtesy and sensitivity are especially important with respect to individuals and
topics dealing with differences of race, color, culture, religion, creed, politics, veteran’s status,
sexual orientation, gender, gender identity and gender expression, age, disability, and
nationalities. Class rosters are provided to the instructor with the student's legal name. I will
gladly honor your request to address you by an alternate name or gender pronoun. Please advise
me of this preference early in the semester so that I may make appropriate changes to my
records. See policies at
http://www.colorado.edu/policies/classbehavior.html and at
Discrimination and Harassment
The University of Colorado Boulder (CU-Boulder) is committed to maintaining a positive
learning, working, and living environment. The University of Colorado does not discriminate on
the basis of race, color, national origin, sex, age, disability, creed, religion, sexual orientation, or
veteran status in admission and access to, and treatment and employment in, its educational
programs and activities. (Regent Law, Article 10, amended 11/8/2001). CU-Boulder will not
tolerate acts of discrimination or harassment based upon Protected Classes or related retaliation
against or by any employee or student. For purposes of this CU-Boulder policy, "Protected
Classes" refers to race, color, national origin, sex, pregnancy, age, disability, creed, religion,
sexual orientation, gender identity, gender expression, or veteran status. Individuals who believe
they have been discriminated against should contact the Office of Discrimination and
Harassment (ODH) at 303-492-2127 or the Office of Student Conduct (OSC) at 303-492-5550.
Information about the ODH, the above referenced policies, and the campus resources available to
assist individuals regarding discrimination or harassment can be obtained at
Academic Integrity
All students of the University of Colorado at Boulder are responsible for knowing and adhering
to the academic integrity policy of this institution. Violations of this policy may include:
cheating, plagiarism, aid of academic dishonesty, fabrication, lying, bribery, and threatening
behavior. All incidents of academic misconduct shall be reported to the Honor Code Council
([email protected]; 303-735-2273). Students who are found to be in violation of the academic
integrity policy will be subject to both academic sanctions from the faculty member and nonacademic sanctions (including but not limited to university probation, suspension, or expulsion).
Other information on the Honor Code can be found at
http://www.colorado.edu/policies/honor.html and at http://honorcode.colorado.edu
Reading List
The list below provides a guide to how to get the most out of your available resources for this
course. Your most directly relevant text will be our lecture notes. They will provide you with
the basics of all of the material that we cover in each class meeting. There are also two books
that I think fit nicely with the applied nature of this course and offer a good complement to our
in-class discussion. They are both relatively inexpensive, and I would recommend them as your
best additional resources for learning the topics we cover. I also strongly recommend having one
or more graduate econometrics textbooks for reference. Finally, we will read a few papers that
actually apply the methods we are discussing. These are listed below in bold. Additional
references that we will probably not have time for are listed in standard font. The links are
active, but you will need to be on-campus or connected through VPN.
Books with an Applied Focus. I highly recommend getting a copy of each of these books, as
they will provide a very useful supplement to my lectures and notes. Angrist and Pischke is
relatively inexpensive (~$25), and I would strongly suggest that each of you get a copy. The
Cameron and Travedi book is great, and it is specifically tailored for people learning STATA. A
good strategy might be to order one for each study group (~$50). As of this writing, they are
currently listed together on Amazon as “Frequently Bought Together”
Angrist and Pischke (2009). Mostly Harmless Econometrics: An Empiricist’s Companion. AP
Cameron and Trivedi (2009). Microeconometrics Using STATA. CT-STATA
Econometrics Reference Books. I am not going to require you to have any particular one of
these. I would recommend that you find at least one of the following books that you find useful
as a reference book. I have tried to include the relevant sections where possible in the main table
Cameron and Trivedi (2005). Microeconometrics: Methods and Applications. CT
Davidson and MacKinnon (2004). Econometric Theory and Methods. DM
Wooldridge (2002). Econometric Analysis of Cross Section and Panel Data. W
Papers. Papers listed in bold are required reading and will be discussed in class during one of
the meetings scheduled for the topic. Exact dates will be announced as we see how we are
progressing. The additional papers listed are for reference for the interested student.
Topics and Readings
Readings marked with a [*] indicate that if I were you, and I had limited time to read nonrequired readings, I would prioritize these.
Introduction and STATA Basics
o Lecture Notes
o [*] CT-STATA Chapter 1
STATA Programming
o Lecture Notes
o [*] CT-STATA Chapter 1.5-1.8, 4
STATA Descriptive Stats, Figures and Tables
o Lecture Notes
o [*] CT-STATA Chapter 2
STATA Data Management
o Lecture Notes
o [*] CT-STATA Chapter 2
Functional Forms
o Lecture Notes
o AP – Chapter 3, various parts
o CT-STATA Chapter 3.3
o CT – Chapter 4.1-4.4
FWL and Multiple Regression
o Lecture Notes
o Lovell(2008) A Simple Proof of the FWL Theorem, Journal of Economic
Education, Vol. 39 No. 1 (Winter 2008)
o DM pp. 68-?
o Zax Textbook, Chapter 12, Section 12.4 pp. 26-35
The RCT/Treatment Effects
o Lecture Notes
o [*] AP – Chapter 2
o W – Chapter 18
Omitted Variable Bias
o Lecture Notes
o [*] AP – Chapter 3.2
o DM – 2.4-2.5
o W – Chapter 4.3
Propensity Score Matching
o Lecture Notes
o Smith and Todd (2005) Does Matching Overcome LaLonde’s Critique of
Nonexperimental Estimators? Journal of Econometrics, Vol 125, No. 1-2, pp.
o Dehejia (2005) Practical Propensity Score Matching: A Reply to Smith and Todd,
Journal of Econometrics, Vol 125, No. 1-2, pp. 355-364
o Smith and Todd (2005) Rejoinder, Journal of Econometrics, Vol 125, No. 1-2, pp.
o [*] CT – Chapter 25.4
o W – Chapter 18.1-18.3
Panel Data – Fixed Effects, etc.
o Lecture Notes
o Ashenfelter and Krueger (1994) Estimates of the Economic Return to
Schooling from a New Sample of Twins, American Economic Review, Vol. 84,
No. 5 (Dec., 1994) pp. 1157-1173
o McKinnish (2008) Panel Data Models and Transitory Fluctuations in the
Explantory Variable. Advances in Econometrics. Vol. 21 2008.
o [*] AP – Chapter 5.1, 5.3, 8.2
o [*] CT-STATA Chapter 8
o CT – Chapter 21
o W – Chapter 10
o Lecture Notes
o Davis (2004) The Effect of Health Risk on Housing Values: Evidence from a
Cancer Cluster. The American Economic Review, Vol. 94, No. 5 (Dec., 2004),
pp. 1693-1704
o Meyer (1995). Natural and Quasi-Experiments in Economics. Journal of
Business and Economic Statistics. Vol. 13, No. 2 pp. 151-161
o [*] AP – Chapter 5.2
o CT – Chapter 22.6
o W – p.130, p. 284
Getting the Standard Errors Right
o Lecture Notes
o Bertrand et. al. (2004) How Much Should We Trust Differences-inDifferences Estimates? Quarterly Journal of Economics, Vol. 119, No. 1,
Pages 249-275
o Moulton (1990) An Illustration of a Pitfall in Estimating the Effects of
Aggregate Variables on Micro Units, Review of Economics and Statistics, Vol.
72, No. 2 (May, 1990), pp. 334-338
o Cameron, Gelbach, and Miller (2006) Robust Inference with Multi-Way
Clustering, NBER Technical Working Paper No. 327
o [*] AP – Chapter 8.2
Instrumental Variables
o Lecture Notes
o Imbens and Angrist (1994) Identification and Estimation of Local Average
Treatment Effects. Econometrica, Vol. 62, No. 2 (Mar., 1994), pp. 467-475
o Bound, Jaeger, and BAPer (1995) Problems with Instrumental Variables
Estimation When the Correlation Between the Instruments and the Endogeneous
Explanatory Variable is Weak. Journal of the American Statistical Association,
Vol. 90, No. 430 (Jun., 1995), pp. 443-450
o Field and Ambrus (2008) Early Marriage, Age of Menarche, and Female
Schooling Attainment in Bangladesh, Journal of Political Economy,Vol 116,
No. 5, pp. 881-930
o [*] AP – Chapter 4
o [*] CT-STATA Chapter 6
o CT – Chapter 4.8-4.9
o DM – Chapter 8
o W – Chapter 5, 18.4
Regression Discontinuity
o Lecture Notes
o Imbens and Lemieux (2008) Regression Discontinuity Designs: A Guide to
Practice. Journal of Econometrics. Volume 142, Issue 2, February 2008, Pages
o Matsudaira (2008) Mandatory Summer School and Student Achievement.
Journal of Econometrics. Volume 142, Issue 2, February 2008, Pages 615-635
o [*] AP – Chapter 6
Binary Dependent Variables
o Lecture Notes
o [*] CT-STATA Chapter 14
o CT – Chapter 14
o W – Chapter 15.1-15.8
Additional Topics2
Event Study Models
o Jacobson, LaLonde, and Sullivan (1993) Earnings Losses of Displaced The
American Economic Review, Vol. 83, No. 4 (Sep., 1993), pp. 685-709
o [*] CT-STATA - Chapter 16
o [*] W – Chapter 17
Propensity Score Reweighting
o DiNardo, Fortin and Lemieux (1996) Labor Market Institutions and the
Distribution of Wages, 1973-1992: A Semiparametric Approach Econometrica,
Vol. 64, No. 5 (Sep., 1996), pp. 1001-1044
Duration Models
I would ideally cover these topics, but in the past, we have not had sufficient time. The citations are provided for
those interested in pursing these topics independently.
[*] Meyer (1990) Unemployment Insurance and Unemployment Spells.
Econometrica, Vol. 58, No. 4 (July 1990), pp. 757-782
o CT – Chapter 17
o W – Chapter 20
Discrete Choice Models
o Train (2009) Discrete Choice Methods with Simulation, Cambridge University
o [*] CT-STATA Chapter 15
o CT – Chapter 15
o W – Chapter 15.9-15.10
Fly UP