Microeconometrics Summer School (2023)

course summary

This course provides the basics for good empirical analysis of panel data. It starts by emphasizing the advantages and limitations of panel data over time series or cross-sectional data. Basic estimation and testing methods for fixed and random effects models are reviewed and illustrated with published empirical applications using Stata and EViews.

Then, endogeneity problems in panel models are examined and instrumental estimation methods for panel variables are reviewed and applied, as well as Hausman-type tests with diverse empirical applications.

The course also includes an introduction to dynamic panel data models, bounded dependent variables and transient panels.

requirements

A solid course in econometrics. Useful knowledge of software such as STATA and/or EViews.

course summary

1. Advantages and limitations of panel data
2. Basic review of panel data methods: estimation and hypothesis testing
3. Simultaneous equations and endogeneity in panel data models
4. Dynamic panel data: Introduction
5. Bounded dependent variables and non-stationary panels

references

Required text (included in the course price):

Baltagi, Badi H., Econometric Analysis of Panel Data, 6. Auflage, Springer, 2021.

https://doi.org/10.1007/978-3-030-53953-5 (23,442 citations from Google Scholar).

Recommended texts:

Baltagi, Badi H., 2009, A Companion to Econometric Analysis of Panel Data, (Chichester, Inglaterra: John Wiley & Sons).

Hsiao, Cheng, 2014, Panel Data Analysis (Cambridge: Cambridge University Press, 3. Aufl.).

Wooldridge, J.M., 2010, Econometric Analysis of Cross-sectional and Panel Data (Cambridge: MIT Press).

Arellano, M., 2003, Paneldatenökonometrie (Oxford: Oxford University Press).

Baltagi, BH, Hrsg. 2015, Oxford Panel Data Handbook, (Oxford: Oxford University Press).

Software-Hardware

STATE*LAPTOP REQUIRED: You must bring your own laptop to attend the hands-on sessions.

about the teacher

Badi Baltagiis a distinguished professor of economics at Syracuse University. He received his Ph.D. in economics from the University of Pennsylvania. He is a senior research associate at the Center for Policy Research at Syracuse University and a research associate at IZA, CESifo, and several other institutions. He was Editor of Economics Letters (2011-2021), Associate Editor, Journal of Econometrics, 1999-2018, Replication Editor, Journal of Applied Econometrics, 2003-2018, Editor, Empirical Economics, 1999-2018. He is currently a member of the Editorial Board of Econometric Reviews. He is a member of the Journal of Econometrics and Econometric Reviews and is the recipient of Econometric Theory's Fine and Plura Scripsit awards. He is also a member of Advances in Econometrics and received the Distinguished Authors Award from the Journal of Applied Econometrics. He is also a Fellow, Founding Member and Past Director of the International Association for Applied Econometrics.

Syracuse University

course summary

The use of research data is becoming common practice among economists and social scientists, both academically and professionally. The main characteristic of these data is that they contain qualitative information, which makes the use of the regression model inadequate when dealing with models in which the dependent variable is a choice (with or without private health insurance) or a state (being or not unemployed). ) or where the dependent variable assumes only non-negative values ​​and a significant percentage of observations is zero (expenditure on certain basic products such as tobacco). On the other hand, the use of information with panel data structures is also becoming more common, as it allows dealing with some limitations of model estimation with cross-sectional data.

In this course we deal with the estimation of these models, paying special attention to the interpretation of the estimates and the limitations of the different models existing in the literature.

The course is divided into four blocks: the first (one and a half sessions) is devoted to discrete selection models, including counting data models; the second (one and a half sessions) on limited models of dependent variables, including typical sample selection problems when dealing with cross-sectional data; the third (one session) is dedicated to duration models (survival analysis) where the dependent variable is the length of stay in a given state (e.g. unemployment) and we are interested in estimating how the duration in a given state , given state changes, affects the probability of exiting that state; and the fourth block an introduction to panel data (static), paying attention to the econometric advantages of this type of information.

The practical sessions are dedicated to estimating these models with STATA and interpreting the results. This is done using real data for different areas: demand analysis, health economics, labor economics, among others. Empirical work is also described and discussed in lectures.

course summary

1. Discrete choice model (I)
a) Binary choice model
b) Multinomial models
c) Models ordered
d) Counting data models​
2. Models for Bounded Dependent Variables (II)
a) Modelo Tobit
b) Pattern selection model
c) Doppelhürdenmodelle
​3. Permanent Models
a) Basic concepts
b) Continuous time models
c) Types of data
4. Dashboard data (static models)
a) Panel data versus cross section
b) Linear model: simple estimators
c) Correlation between individual effects and regressors

references

• Cameron, AC and Trivedi, P.K., Microeconometrics. Methods and Applications, Cambridge University Press, 2005
• Cameron, AC and Trivedi, P.K., Microeconometrics with STATA, STATA Press, 2010
• Wooldridge, J.M., Econometric Analysis of Panel and Cross-Section Data, MIT Press, 2010

Software-Hardware

STATE*LAPTOP REQUIRED: You must bring your own laptop to attend the hands-on sessions.

about the teacher

Jaume Garcia-VillarHe is Professor of Economics at Pompeu Fabra University and Associate Professor at BSE. He received his PhD from the London School of Economics and Political Science in 1985. From 2008 to 2011 he was President of the Spanish National Institute of Statistics.

course summary

The main challenge for policy evaluation is to establish a causal relationship between interventions and outcomes. The objective of this course is to present the main econometric approaches used in the evaluation of public policies: random evaluations, natural experiments, discontinuous regression design, observable selection, differences in differences and synthetic control methods. The course presents the strengths and weaknesses of each approach in terms of internal and external validity. During the morning sessions, each approach will be presented and illustrated with concrete examples from the areas of labor economics, health economics and educational economics. In the afternoon sessions, we will replicate the results of one of the major published studies for each assessment approach in Stata. In preparation, students receive the relevant data and codes in advance.

course summary

• Randomized evaluations (experiments)
• Natural Experiments
• regression discontinuity designs
• Selection by observables (linear regression, matching)
• Differences in differences, event studies, synthetic control methods

references

• Abadie, A. und M. D. Cattaneo, 2018, „Econometric Methods for Program Evaluation“,Annual Economic Review, 10: 465-503.
• Imbens, G.W. and J.M. Wooldridge, 2009, „Recent Developments in Program Evaluation Econometrics“,business literature magazine, 47 (1): 5-86.

Software-Hardware

STATE*LAPTOP REQUIRED: You must bring your own laptop to attend the hands-on sessions.

about the teacher

albrecht shineHe is an associate professor at the Universitat Pompeu Fabra, an associate professor at the Barcelona School of Economics and a researcher at IPEG. He received his PhD from University College London in 2008. His research interests include labor economics, migration economics and microeconometrics. His work was published inAmerican Economic Review, to dieeconomics review, it's himlabor economy magazine, between others.

course summary

This course deals with methods and applications of discrete choice dynamic structural models in economics. The methods we will discuss are commonly used by researchers interested in labor economics, industrial organization, quantitative macroeconomics, development economics, and many other areas of applied microscience. The course focuses on the estimation of discrete choice dynamic structural models that allow modeling the behavior of predictive agents that make discrete decisions. These models represent a useful tool for policy evaluation and an interesting complement to reduced-form approaches, in particular, they have the advantage of allowing ex-ante evaluation of policies and conferring external validity to the conclusion, depending on the assumptions of the model. They also have the advantage of providing a close link between economic theory and empiricism and being able to make inferences about a model's predictions.

The course is organized into three blocks. In the first (3 hours) we present the basic framework and discuss standard estimation techniques based on maximum likelihood, where dynamic programming problems are numerically solved. The second (5 hours) covers an alternative set of estimation methods grouped under the label "Conditional Choice Probability" estimation methods that avoid the need to solve for value functions in estimation. Finally, the third (2 hours) examines the additional complications posed by dynamic problems involving game theory reactions to the decisions of other agents. The theory is complemented with examples of work published in the literature.

The practical sessions will be dedicated to the numerical solution and estimation of these models using STATA (MATA), using simple empirically relevant applications and giving instructions on how these simple applications can be extended to rich and complete environments like the ones presented in the described classes. in published articles are appreciated.

Both theoretical and practical sessions are designed to be interactive. Depending on class size, split sessions will be considered to encourage interaction, particularly in practical sessions.

requirements

Knowledge of econometrics (at the level of Wooldridge, 2002) and dynamic programming (at the level of Chapters 1 and 2 of Adda and Cooper, 2003) are strongly recommended.

course summary

1. Complete Solutions Maximum Likelihood Approaches
   The introduction
   B. Structure: Conditional Independence
   C. Motivational Example: Rust Engine Replacement Model
   D. Estimation using complete solution techniques
   my. Extensions: heterogeneity and equilibrium not observed
2. Estimation of conditional vote probability (PCC).
   A. Representation of the conditional value function
   B. finite dependency
   C. Estimation methods
   D. Aguirregabiria and Mira's iterative approach
   my. Extensions: heterogeneity and equilibrium not observed
3. An introduction to dynamic discrete games

references

1. Adda, J. and R. W. Cooper (2003), Dynamic Economics: Quantitative Methods and Applications. Die MIT-Press.
2. Altug, S. und R. A. Miller (1998), „The Effect of Work Experience on Women's Wages and Labor Supply“, Review of Economic Studies, 65, 45-85.
3. Aguirregabiria, V. und P. Mira (2002), „Swapping the Nested Fixed Point Algorithm: A Class of Estimators for Discrete Markov Decision Models“, Econometrica, 70, 1519-1543.
4. Aguirregabiria, V. und P. Mira (2007), „Sequential estimation of discrete dynamic games“, Econometrica, 75, 1-53.
5. Aguirregabiria, V. und P. Mira (2010), „Dynamic Discrete Choice Structural Models: A Survey“, Journal of Econometrics, 156: 38-67
6. Arcidiacono, P. und P. B. Ellickson (2011), „Practical Methods for Estimation of Dynamic Discrete Choice Models“, Annual Review of Economics, 3, 363-394.
7. Arcidiacono, P. und R. A. Miller (2011), „Conditional choice probability estimation of dynamic discrete choice models with unobserved heterogeneity“, Econometrica, 79, 1823-1867.
8. Berndt, EK, BH Hall, RE Hall, J A Hausman (1974). "Estimation and Inference in Nonlinear Structural Models," Annals of Economic and Social Measurement 3, 653-665.
9. Eckstein, Z. und K. Wolpin (1989), „The specification and estimation of discrete choice dynamic stochastic models: a survey“, Journal of Human Resources, 24: 562-598
10. Hong H and M Shum (1998) Structural estimation of auction models in Patrone F, García-Jurado I, Tijs S (eds) Game Practice: Contributions from Applied Game Theory. Decision Theory Library (Series C: Game Theory, Mathematical Programming, and Operations Research), Volume 23. Springer, Boston, MA.
11. Hotz, V. J. e R. A. Miller (1993), "Conditional Choice Probabilities and Estimation of Dynamic Structural Models", Review of Economic Studies, 60, 497-529.
12. Keane, M. P. und K. I. Wolpin (1997), „The Career Decisions of Young Men“, Journal of Political Economy, 105, 473-522.
13. Lee, D. und K. I. Wolpin (2006), „Intersectoral labor mobility and the growth of the service sector“, Econometrica, 74, 1-46.
14. Llull, J. (2018), „Immigration, wages and education: a structural model of labor market equilibrium“, Journal of Economic Studies, 85, 1852-1896.
15. Llull, J. (2018), „Selective Immigration Policies and the US Labor Market“, mimeo, MOVE, Universitat Autònoma de Barcelona, ​​​​​​​and BSE.
16. Miller, R. A. (1997), „Estimation of dynamic optimization models with microeconomic data“, em M. Pesaran und P. Schmidt (Hrsg.), Handbook of Applied Econometrics, vol. 3, não. 2, arte. 246-299
17. Todd, P. e K. Wolpin (2006), “Evaluating the Impact of School Subsid Programs in Mexico: Using a Social Experiment to Validate a Dynamic Behavioral Model of School Enrollment and Child Fertility,” American Economic Review, 96, 1384-1417 .
18. Rust, J. (1987), "Ideal Replacement of GMC Bus Engines: An Empirical Model by Harold Zürcher", Econometrica, 55, 999-1033.
19. Rust, J. (1994), "Structural Estimation of Markov Decision Processes", em R. E. Engle und D. L. McFadden (Hrsg.), Handbook of Econometrics, vol. 3, nº 4, cap. 51.

Software-Hardware

STATE*LAPTOP REQUIRED: You must bring your own laptop to attend the hands-on sessions.

about the teacher

John LlullHe is Director of MOVE, Associate Professor of Economics at the Autonomous University of Barcelona and Associate Research Professor at the Barcelona School of Economics. He is also a CReAM External Fellow (UCL).

Professor Llull's research focuses on labor economics and, in particular, immigration, internal migration, labor mobility, inequality, human capital, family economics and health. His main research is usually estimating dynamic discrete-choice equilibrium models, but several of his papers also use reduced-form approaches. His work was published ineconomics reviewS,HR Diary,it's himeuropean economic report, between others.

Professor Llull is the data editor for thebusiness diary, member of the editorial boardeconomics review, guest editor of a special issue on "Economics of migration: impacts on the labor market and migration policies" bylabor economy, o Journal of the European Association of Labor Economists, e co-editor doseries--- Magazine of the Spanish Economic Association.

course summary

This course covers difference-in-differences (DiD) estimators for policy analysis using panel data. Emphasis is placed on using and combining Stata's built-in commands together with newly written user commands to allow for simple estimation methods, robust inference, and flexibility in treatment effect patterns. We will see how the usual two-way fixed effects approach can be extended in the case of graded intervention to allow for heterogeneous treatment effects across cohort and time.

In addition, we will see how covariates are included in the analysis in all settings. Simple solutions for violations of parallel trend assumptions are also covered. Problems and approaches with skewed panels and statistical inference problems with too few treatment or control groups are also addressed.

The course ends with a discussion of my recent research on step-by-step designs using non-linear models, with an emphasis on binary, fractional, and non-negative responses.

requirements

Participants should have a good working knowledge of ordinary least squares estimation, fixed effects estimation, and basic nonlinear models such as logit, probit, and exponential conditional means. Sufficient information is provided in my introductory econometrics book, Introductory Econometrics: A Modern Approach, 7e, Cengage, 2020. My book Econometric Analysis of Cross Section and Panel Data, 2e, MIT Press, 2010 covers the reference material at a superior level . An optional approximately 90-minute video provides background information on possible outcomes and estimates of treatment effects. This material is applied in DiD contexts.

course summary

1. Introduction. Differences in the differences of two periods. parallel trends. Control of covariates by propensity score and regression fit methods.
2. More than two periods. Fixed bidirectional effects. Heterogeneous effects of treatment. Control for regression fit covariates and estimates of treatment effects.
3. Stepped Interventions, I. Combined OLS and Extended TWFE. Accumulation. Test and fix parallel trend violations. Unbalanced panels.
4. Interventions by stages, II. Evolutionary process. Long differentiation. Propensity score and doubly robust methods. imputation methods. Clustering by inference with few treated units.
5. Non-linear DiD. Binary, fractional and non-negative answers. Pooled near-maximum probability estimate. Fixed poison effects.

references

For part of the course, I will use my recent work papers Two-Way Fixed Effects, Mundlak Two-Way Regression, and Difference-in-Differences Estimators, and Simple Approaches to Differences in Nonlinear Differences with Panel Data.

Software-Hardware

STATE*LAPTOP REQUIRED: You must bring your own laptop to attend the hands-on sessions.

about the teacher

Jeffrey M. Wooldridgeis Distinguished University Professor of Economics at Michigan State University, where he has taught since 1991. From 1986 to 1991 Dr. Wooldridge Assistant Professor of Economics at the Massachusetts Institute of Technology. He received his Bachelor of Arts degree with a concentration in Computer Science and Economics from the University of California, Berkeley, in 1982 with High Honors in General Science. He received his Ph.D. in economics from the University of California, San Diego, in 1986.

dr. Wooldridge has published a large number of articles in internationally recognized journals, as well as numerous book chapters, including articles in the Handbook of Econometrics and the Handbook of Applied Econometrics. He is the author of Introductory Econometrics: A Modern Approach (South-Western, 4e, 2009) and Econometric Analysis of Cross Section and Panel Data (MIT Press, 2002).

dr. Wooldridge is a member of the Econometric Society and the Journal of Econometrics. His other awards include the Plura Scripset Award for Econometric Theory, the Sir Richard Stone Award from the Journal of Applied Econometrics, and three Professor of the Year awards from the MIT Graduate Economics Association. He was also an Alfred P. Sloan Research Fellow.

dr. Wooldridge currently serves on the editorial board of the Stata Journal. He has been editor of the Journal of Economic Literature, Journal of Business and Economic Statistics, associate editor of Econometric Theory, associate editor for econometrics of Economics Letters, and associate editor of the Journal of Econometrics and Review of Economics and Statistics.

course summary

This course provides up-to-date information on dynamic panel data models, discrete choice panel data models, as well as censored panel data models and subject choice dynamic panel data model estimation. In addition to an overview of relevant theory, the focus of the course is on the practical application of these models to different data contexts: small T - large N, unbalanced panels, rotating panels and pseudo-panels constructed from cohort data.

The practical exercises provide guidance to the public on how to deal with these models, in particular, how to formulate the constraints and contrasts for a proper specification of the model to be estimated. All practical exercises are solved with STATA.

requirements

Some prior knowledge in econometrics is strongly recommended.

course summary

1. Review of linear panel data models and introduction to dynamic panel data models
2. Dynamic dashboard data and extensions
3. Censored panel data models
4. Sample selection panel data templates. new developments
5. Dashboard data templates with discrete options: static and dynamic

basic references

• Arellano, M. (1992), „Discrete choice for panel data“, Economic Research, 2003.
• Arellano, M. und O. Bover (1995), „Another look at estimating instrumental variables from error component models“, Journal of Econometrics 68, 29-51.
• Arellano, M. and R. Carrasco (2003), “Binary choice models with standard variables”, Journal of Econometrics, 115, 155-165.
• Arellano, M. und B. Honoré (2003), ”Panel Data Models, some new developments”, Journal of Econometrics, 115, 155-165.
• Baltagi, BH (1995), Econometric analysis of panel data, John Wiley. 4th edition 2008.
• Blundell, R. e S. Bond (1998), "Initial Conditions and Moment Constraints in Dynamic Panel Data Models," Journal of Econometrics 87, 115-143.
• Steve Bond, 2002, „Dynamic Panel Data Models: A Guide to Microdata Methods and Practices“, CeMMAP Working Papers CWP09/02, Center for Microdata Methods and Practices, Institute for Fiscal Studies.
• Collado, MD (1997), "Estimating Dynamic Time Series Models of Independent Cross Sections", Journal of Econometrics, 82, 37–62.
• Deaton, A. 1985. Time series cross-section panel data. Journal of Econometrics 30, 109-126.
• Dustman, C. und Rochina-Barrachina, M. R. (2007) "Selection Correction in Panel Data Models: An Application to Labor Supply and Wages" Econometrics Journal (2007), Band 10, S. 263293. doi: 10.1111 / j.1368- 423X.2007.00208.x
• Jimenez-Martin, Labeaga and Rochina-Barrachina (2009), "Comparison of estimators in dynamic panel data sample selection and switching models", Mimeo, presented at the CambridgePD conference.
• Sergi Jiménez-Martín and José María Labeaga, 2016. "Monte Carlo Evidence for Estimating Sampling Models for AR(1) Panel Data", Working Papers 2016-01, FEDEA.
• Hansen, LP (1982), "Large Sample Properties of the Generalized Method of Moment Estimators", Econometrica 50, 1029-1054.
• Jiménez-Martín, S. (1999), "Controlling the Endogeneity of Strike Variables in the Estimation of Payroll Equations", Journal of Labor Economics, 17, 587-606
• Jiménez-Martín, S. & J.M. Labeaga & M.al Sadoon, 2020. "Consistent estimation of sample selection models for panel data", Working Papers 2020-06, FEDEA.
• Jones, AM and JM Labeaga (2003), “Individual heterogeneity and censorship in panel data estimates of tobacco expenditures”, Journal of Applied Econometrics, 18, 157-177.
• Kyriazidou, E. (1997), „Estimation of a panel data sample selection model“, Econometrica 65, 1335-1355
• Kyriazidou, E. (2001), „Estimating Dynamic Sample Selection Models from Panel Data“, Review of Economic Studies 68, 543-572.
• Labeaga, JM. (2001), “A Rational Model of Addiction with Two Hurdles and Heterogeneity: Estimating Tobacco Demand,” Journal of Econometrics, 93, 49-72.
• Labeaga, JM. (2001), “Efficiency comparisons in dynamic panel data models with bounded dependent variables”, WP, UNED, Madrid.
• Nickell, S. (1981), „Biases in Dynamic Models with Fixed Effects“, Econometrica 49, 1417-1426.
• Semykina A. and Jeffrey M. Wooldridge „Estimating a Dynamic Panel Data Model with Sample Selection“, April 28, 2009
• Semykina, A. und Wooldrigde, J. M. (2005), „Model estimation from panel data in the presence of endogeneity and selection: theory and application“, mimeographed.
• Windmeijer, Frank, 2005. "A Finite Sample Correction for the Variance of Linearly Efficient Two-Stage GMM Estimators," Journal of Econometrics, Elsevier, vol. 126(1), pp. 25-51, maio.
• Wooldridge, J. M. "Selection Corrections for Panel Data Under Conditional Mean Independence Assumptions," Journal of Econometrics, 68 (1995): 115-132.
• Majid M. Al-Sadoon & Sergi Jiménez-Martín & José M. Labeaga, 2019. „Simple methods for consistent estimation of dynamic panel data sample selection models“, Working Papers 1069, Barcelona School of Economics.

Software-Hardware

STATE*PORTABLE REQUIRED:You must bring your own laptop to participate in the practical exercises.

about the teacher

Sergi Jiménez-MartínHe is Professor of Economics at Pompeu Fabra University and Associate Professor at BSE. Doctorate from Pompeu Fabra University in 1994, he is currently Professor of Health Economics and Life Habits at FEDEA-La Caixa, Associate Editor of Empirical Economics and member of the Scientific Council of Applied Economic Perspectives and Policies and Economic Notebooks of ICE.