HMP 826
Applied Econometrics in Health Services Research
Winter 2018
PROFESSOR
Edward C. Norton, Ph.D.
Professor of Health Management and Policy
Professor of Economics
M3108 SPH II
ecnorton@umich.edu
734-615-5738
Office Hours: by appointment
CLASS SCHEDULE
Main lectures
Problem sessions
Mondays and Wednesdays 8:3010:00 am in room 2750 SPH I.
Thursdays 1:302:30 pm in M1138 SPH II.
Fridays 11:0012:00 am in M4318 SPH II.
COURSE DESCRIPTION
This course is an introduction to applied econometrics in health services research
using maximum likelihood estimation, the mathematical framework for the
course. We will study econometric models in which the dependent variable is not
continuous, including logit and probit, multinomial logit, ordered logit and probit,
tobit, selection, two-part, and count models. The latter part of the course provides
an integrated approach to specification tests. This course teaches how to choose
the appropriate model, check the model specification, and interpret and present
the results. Students will apply these techniques in weekly problem sets and an
empirical term paper.
COURSE MATERIALS
The lecture notes, the basis of this course, are available at Dollar Bill, 611 Church St.
 Norton, EC. 2016. HMP 826 Class Notes: Applied Econometrics in Health Services
Research. [The coursepack is printed in two parts.]
 Deb, P., E.C. Norton, and W.G. Manning. 2017. Health Econometrics Using Stata.
Stata Press.
 Additional readings are on Canvas.
Other resources may also be helpful.
 Cameron, AC and PK Trivedi. 2009. Microeconometrics Using Stata.
 StataCorp. 2017. Stata Statistical Software: Release 15. College Station, TX:
StataCorp LP.
PREREQUISITES
Economics 571 or equivalent, or permission by instructor.
COURSE GOALS
The goals of the course are related to the third CEPH foundational learning
objective. The goals of this course are to understand how to
1. estimate and interpret logit, probit, and other maximum likelihood models;
2. build appropriate models in an MLE framework; and
3. test the model specifications.
COURSE COMPETENCIES
The course competencies are related to the second, third, and fourth CEPH
foundational competencies for evidence-based approaches to public health.
COURSE REQUIREMENTS, ASSIGNMENTS, AND GRADING
The course grade will be a weighted average of performance in four areas. There
will be weekly problem sets, a midterm exam, a final exam, and an empirical
paper due in class on the last day of class.
Grading:
Problem sets
Midterm exam
Final exam
Empirical paper
Total
10%
30%
30%
30%
100%
Problem sets, which are due on Wednesdays, will include a mix of written
problems and computer assignments. Turn in a hard copy. The following simple
grading scheme will be used for each problem: 2 points if entirely correct; 1 point
if close to correct; 0 points if not close to correct.
The empirical paper allows the students to demonstrate the skills learned
throughout the semester. Students should start looking now for an interesting
empirical question to answer and a data set that can be used to answer that
question. The empirical paper is due in class on the last day of class. It will be
given a letter grade with comments.
CLASSROOM EXPECTATIONS AND ETIQUETTE
Students are allowed and encouraged to collaborate on assignments (but not
exams), unless instructed otherwise. To emphasize the importance of integrity
2
and intellectual property in the profession, on each assignment please list your
collaborators. Academic misconduct will not be tolerated.
DIVERSITY, EQUITY, AND INCLUSION
SPH is committed to creating classroom environments that are supportive of
diversity, equity and inclusion.
ACADEMIC INTEGRITY
The faculty and staff of the School of Public Health believe that the conduct of a
student registered or taking courses in the School should be consistent with that of
a professional person. Courtesy, honesty, and respect should be shown by
students toward faculty members, guest lecturers, administrative support staff,
community partners, and fellow students. Similarly, students should expect
faculty to treat them fairly, showing respect for their ideas and opinions and
striving to help them achieve maximum benefits from their experience in the
School.
Student academic misconduct refers to behavior that may include plagiarism,
cheating, fabrication, falsification of records or official documents, intentional
misuse of equipment or materials (including library materials), and aiding and
abetting the perpetration of such acts. Please visit https://sph.umich.edu/studentresources/mph-mhsa.html for the full Policy on Student Academic Conduct
Standards and Procedures.
STUDENT WRITING LAB
The SPH Writing Lab is located in 5025 SPH II and offers writing support to all
SPH students for course papers, manuscripts, grant proposals, dissertations,
personal statements, and all other academic writing tasks. The Lab can also help
answer questions on academic integrity. To learn more or make an appointment,
please visit the SPH writing lab website.
STUDENT WELL-BEING
SPH faculty and staff believe it is important to support the physical and emotional
well-being of our students. If you have a physical or mental health issue that is
affecting your performance or participation in any course, and/or if you need help
connecting with University services, please contact the instructor or the SPH
Office for Student Engagement and Practice. Please visit
https://sph.umich.edu/student-life/wellness.html for information on wellness
resources available to you.
3
STUDENT ACCOMMODATIONS
Students should speak with their instructors before or during the first week of
classes regarding any special needs. Students can also visit the SPH Office for
Student Engagement and Practice for assistance in coordinating communications
around accommodations. Students seeking academic accommodations should
register with Services for Students with Disabilities (SSD). SSD arranges
reasonable and appropriate academic accommodations for students with
disabilities. Please visit https://ssd.umich.edu/topic/our-services for more
information on student accommodations.
Students who expect to miss classes, examinations, or other assignments as a
consequence of their religious observance shall be provided with a reasonable
alternative opportunity to complete such academic responsibilities. It is the
obligation of students to provide faculty with reasonable notice of the dates of
religious holidays on which they will be absent. Please visit
http://www.provost.umich.edu/calendar/ for the complete University policy.
4
COURSE TOPICS
MAXIMUM LIKELIHOOD ESTIMATION
Chapter 1
Chapter 2
Introduction and math review
Maximum likelihood estimation
[DNM Chapter 3]
LOGIT AND PROBIT MODELS
Chapter 3
Chapter 4
Chapter 5
Chapter 6
Chapter 7
Chapter 8
Chapter 9
Logit and probit models
Logit and probit estimation
Logit and probit interpretation
Interaction terms in non-linear models
Log odds and ends
Logit and probit models with panel data
Logit and probit models with endogeneity
[DNM Chapter 2]
OTHER CATEGORICAL DATA MODELS
Chapter 10
Chapter 11
Chapter 12
Chapter 24
Chapter 13
Chapter 14
Chapter 15
Chapter 16
Multinomial logit
Extensions to multinomial logit
Ordered logit and ordered probit
Bootstrapping
Tobit models
Selection models
Two-part models
Count models
[DNM Chapter 11]
[DNM Chapter 7]
[DNM Chapter 7]
[DNM Chapter 7]
[DNM Chapter 8]
SPECIFICATION TESTS
Chapter 17
Chapter 18
Chapter 19
Chapter 20
Chapter 21
Chapter 22
Chapter 23
Introduction to specification tests
Mathematics of specification tests
LR tests
Wald tests
LM tests
Hausman tests
Endogeneity tests for binary outcomes
5
[DNM Chapter 4]
[DNM Chapter 10]
COURSE SCHEDULE: FIRST HALF
January 3
January 5
Chapter 1
Introduction and math review
No problem session
January 8
January 10
January 12
Chapter 2
Maximum likelihood estimation
Chapter 3
Logit and probit models
Problem session
January 15
January 17
January 19
Martin Luther King, Jr. Day, class will not be held
Chapter 4
Logit and probit estimation
Problem session
January 22
January 24
January 26
Chapter 5
Logit and probit interpretation
Chapter 6
Interaction terms in non-linear models
Problem session
January 29
January 31
February 2
Chapter 7
Log odds and ends
Chapter 7
Log odds and ends
Problem session
February 5
February 7
February 9
Chapter 8
Logit and probit models with panel data
Chapter 9
Logit and probit models with endogeneity
Problem session
February 12
February 14
February 16
Chapter 10
Multinomial logit
Chapter 12
Ordered logit and ordered probit
Problem session
February 19
February 21
February 23
Chapter 24
Midterm
No session
February 26 – March 2
Bootstrapping
Spring Break, classes will not be held
6
COURSE SCHEDULE: SECOND HALF
March 5
March 7
March 9
Chapter 13
Tobit models
Chapter 14
Selection models
Problem session
March 12
March 14
March 16
Chapter 15
Two-part models
Chapter 15
Two-part models
Problem session
March 19
March 21
March 23
Chapter 16
Count models
Chapter 16
Count models
Problem session
March 26
March 38
March 30
Chapter 16
Count models: two-part and zero-inflated
Chapter 17
Introduction to specification tests
Problem session
April 2
April 4
April 6
Chapter 18
Mathematics of specification tests
Chapter 19
LR tests
Problem session
April 9
April 11
April 13
Chapter 20
Wald tests
Chapter 21
LM tests
Problem session
April 16
Chapter 22
April 25
Final Exam, 8:00 am – 10:00 am
Hausman tests
Empirical papers are due in class
7
COURSE TOPICS AND READING LIST
Chapters 12. Maximum Likelihood Estimation
Introduction and math review
Chapter 1
 Deb, Norton, and Manning. Chapter 3. (To introduce data for problem sets.)
 Duggan M. and S.D. Levitt. 2002. “Winning isn’t everything: Corruption in sumo
wrestling.” American Economic Review 92(5):15941605.
Maximum likelihood estimation
Chapter 2
Chapters 39. Logit and Probit Models
Logit and probit models
Chapter 3
Logit and probit estimation
Chapter 4
Logit and probit interpretation
Chapter 5
 Deb, Norton, and Manning. Section 2.5.
 Buchmueller, T.C., A.T. Lo Sasso, I. Lurie, and S. Dolfin. 2007. “Immigrants and
employer sponsored health insurance.” Health Services Research 42(1):286310.
(Probit)
Interaction terms
Chapter 6
 Ai, C. and E.C. Norton. 2003. “Interaction Terms in Logit and Probit Models.”
Economics Letters 80(1):123129.
 Greene, W. 2010. “Testing hypotheses about interaction terms in nonlinear models.”
Economics Letters 107(2):291296.
 Karaca-Mandic, P., E.C. Norton, and B.E. Dowd. 2012. “Interaction terms in nonlinear models.” Health Services Research 47(1):255–274.
 Norton, E.C., H. Wang, and C. Ai. 2004. “Computing interaction effects and standard
errors in logit and probit models.” Stata Journal 4(2):103116.
 Puhani. 2012. “The treatment effect, the cross difference, and the interaction term in
non-linear “Difference-in-differences” models.” Economics Letters 115(1):8587.
Log odds and ends
Chapter 7
8
 Kleinman, L.C. and E.C. Norton. 2009. “What’s the Risk? A simple approach for
estimating adjusted risk ratios from nonlinear models including logistic regression.”
Health Services Research 44(1):288302.
 Mood C. 2010. Logistic Regression: Why We Cannot Do What We Think We Can
Do, and What We Can Do About It. European Sociological Review 26(1): 67–82.
 Norton, E.C. and B.E. Dowd. 2018. “Log odds and the interpretation of logit models.”
Health Services Research 53(2):859878.
 Norton, E.C., M.M. Miller, L.C. Kleinman. 2013. “Computing adjusted risk ratios and
risk differences in Stata” Stata Journal 13(3):492509.
Logit and probit models with panel data
Chapter 8
 Norton, E.C. and C.H. Van Houtven. 2006. “Inter-vivos Transfers and Exchange.”
Southern Economic Journal. 73(1):157172. (Chamberlain conditional logit)
 Zarkin, G.A., L.J. Dunlap, J.W. Bray, W.M. Wechsberg. 2002. “The effect of
treatment completion and length of stay on employment and crime in outpatient drugfree treatment.” Journal of Substance Abuse Treatment 23:261271.
Logit and probit models with endogeneity Chapter 9
 Angrist, JD and Krueger, AB. 2001. Instrumental variables and the search for
identification: From supply and demand to natural experiments. Journal of Economic
Perspectives 15(4):6985.
 Basu, A and N Coe. 2015. 2SLS vs. 2SRI: Appropriate methods for rare outcomes
and/or rare exposures. University of Washington working paper.
 Murray, M.P. 2006. “Avoiding Invalid Instruments and Coping with Weak
Instruments.” The Journal of Economic Perspectives 20(4):111132.
 Terza, J.V., A. Basu, and P.J. Rathouz. 2008. “Two-stage residual inclusion
estimation: Addressing endogeneity in health econometric modeling.” Journal of
Health Economics 27(3):531543.
 Terza, J.V., W.D. Bradford, and C.E. Dismuke. 2008. “The use of linear instrumental
variables methods in health services research and health economics: A cautionary
note.” Health Services Research 43(3):11021120.
Chapters 1016. Other Categorical Data Models
Multinomial logit
Chapter 10
 Norton, E.C. and D.O. Staiger. 1994. “How hospital ownership affects access to care
for the uninsured.” RAND Journal of Economics 25 (1): 171185. (Multinomial logit)
Extensions to multinomial logit
Chapter 11
9
 Tay, Abigail. 2003. Assessing Competition in Hospital Care Markets: The Importance
of Accounting for Quality Differentiation. RAND Journal of Economics
34(4):786815. (McFadden conditional logit)
Ordered logit and ordered probit
Chapter 12
 Chaloupka, F.J. and H. Wechsler. 1997. “Price, tobacco control policies and smoking
among young adults.” Journal of Health Economics 16(3):359373. (Ordered probit)
Bootstrapping
Chapter 24
 Deb, Norton, and Manning. Section 11.3.2.
 Domino, M.E., C. Humble, W.W. Lawrence, and S. Wegner. 2009. “Enhancing the
medical homes model for children with asthma.” Medical Care 47(11):11131120.
 Dowd, B.E., W.H. Greene, E.C. Norton. 2014. “Computation of Standard Errors”
Health Services Research 49(2):731–750
Tobit models
Chapter 13
 Deb, Norton, and Manning. Section 7.6.
 Holmes, A.M., and P. Deb. 1998. “Provider choice and use of mental health care:
Implications for gatekeeper models.” Health Services Research 33(5):12631284.
Selection models
Chapter 14
 Deb, Norton, and Manning. Sections 7.1, 7.3, 7.5.
 Bundorf, M.K. 2002. “Employee demand for health insurance and employer health
plan choices.” Journal of Health Economics 21(1):6588. (Heckman selection)
 Carmichael, F. and S. Charles. 2003. “The opportunity costs of informal care: does
gender matter?” Journal of Health Economics 22(5):781803. (Heckman selection)
 Puhani, PA. 2002. The Heckman correction for sample selection and its critique.
Journal of Economic Surveys 14(1):53–68.
Two-part models
Chapter 15
 Deb, Norton, and Manning. Chapter 7.
 Belotti, F., P. Deb, W.G. Manning, E.C. Norton. 2015. “twopm: Two-part models.”
Stata Journal 15(1):320.
 Stuart, B.C., J.A. Doshi, and J.V. Terza. 2009. “Assessing the impact of drug use on
hospital costs.” Health Services Research 44(1):128144.
 Van Houtven, CH and EC Norton. 2004. Informal care and health care use of older
adults. Journal of Health Economics 23(6):11591180.
Count models
Chapter 16
10
 Deb, Norton, and Manning. Chapter 8.
 Morgan, P.A., N.D. Shah, J.S. Kaufman, and M.A. Albanese. 2008. “Impact of
physician assistant care on office visit resource use in the United States.” Health
Services Research 43(5):19061922.
Chapters 1723. Specification Tests
Specification tests
Chapter 17 & 18
 Deb, Norton, and Manning. Chapter 4.
LR tests
Chapter 19
 Cramer, JS and G Ridder. 1991. “Pooling states in the multinomial logit model.”
Journal of Econometrics. 47(2/3):267-272.
Wald tests
Chapter 20
LM tests
Chapter 21
 MacKinnon, JG. 1992. “Model Specification Tests and Artificial Regressions.”
Journal of Economic Literature. Vol. XXX:102146.
Hausman tests
Chapter 22
Endogeneity tests for binary outcomes
Chapter 23
 Deb, Norton, and Manning. Section 10.2.3.
 Blundell, R.W. and R.J. Smith. 1989. “Estimation in a class of simultaneous equation
limited dependent variable models.” Review of Economic Studies 56 (1): 3757.
 Bolen, K.A., D.K. Guilkey, and T.A. Mroz. 1995. “Binary outcomes and endogenous
explanatory variables: Tests and solutions with an application to the demand for
contraceptive use in Tunisia.” Demography 32(1):111131.
 Rivers, D. and Q.H. Vuong. 1988. “Limited information estimators and exogeneity
tests for simultaneous probit models.” Journal of Econometrics 39 (3): 347366.
 Smith, R.J. and R.W. Blundell. 1986. “An exogeneity test for a simultaneous equation
Tobit model with an application to labor supply.” Econometrica 54 (3): 679685.
11
ADDITIONAL REFERENCES
Additional references can be found online, including at the Stata Bookstore (www.stata.com).
The following textbooks are focused on topics covered in this class.
 Deb, P., E.C. Norton, and W.G. Manning. 2017. Health Econometrics Using Stata.
Stata Press.
 Greene, W.H. and D.A. Hensher. 2010. Modeling Ordered Choices. Cambridge
University Press.
 Long, J.S. and J. Freese. 2001. Regression Models for Categorical Dependent
Variables using Stata. Stata Press.
The following textbooks are good general textbooks in econometrics.
 Angrist, J.D. and J.-S. Pischke. 2009. Mostly Harmless Econometrics: An Empiricist’s
Companion. Princeton: Princeton University Press.
 Cameron, A.C. and P.K. Trivedi. 2005. Microeconometrics: Methods and
Applications. New York: Cambridge University Press.
 Greene, W.H. 2011. Econometric Analysis. Seventh edition. Prentice-Hall.
 Kennedy, P. 2008. A Guide to Econometrics. Sixth edition. Blackwell Publishing.
 Wooldridge, J.M. 2009. Introductory Econometrics: A Modern Approach. Fourth
edition. South-Western College Publishing.
 Wooldridge, J.M. 2010. Econometric Analysis of Cross Section and Panel Data.
Second edition. MIT Press.
Additional interesting readings.
 Siegfried, JJ. 1970. A First Lesson in Econometrics. Journal of Political Economy
78(6):13781379.
 Fuchs, VR. 1986. Physician-induced demand: A parable. Journal of Health Economics
5(4):367.
 Kennedy, P. 2002. Sinning in the Basement: What are the Rules? The Ten
Commandments of Applied Econometrics. Journal of Economic Surveys 16(4):569–
589.
 Smith, G. and J. Pell. 2003. Parachute Use to Prevent Death and Major Trauma
Related to Gravitational Challenge: Systematic Review of Randomized Control Trials.
British Medical Journal 327:1459–1461.
 Efron, B. 2013. A 250-year argument: Belief, behavior, and the bootstrap. Bulletin of
the American Mathematical Society 50(1):129–146.
12