Topics in Econometrics: Online Syllabus

Course Description

This course covers advanced topics in econometrics with a particular focus on topics in causal inference for applications in economics. The course is designed for students interested in pursuing research in applied and theoretical econometrics, statistics or computer science as well as for students with a solid background in econometrics or statistics interested in learning recent advances in causal inference for applied economic research. The first part presents topics in causal inference, starting from selection on observables, non-compliance, and then focusing on experimental design, and inference in experiments. The second part covers most advanced topics for semi-parametric estimation, heterogeneous treatment effects and policy learning, instrumental variables, panel data models, spillover effects. The class concludes with an overview of recent research in econometrics on the design of scientific communication. Readings for each topic will include theoretical papers in econometrics and papers in applied economic research that provide empirical examples of the method.

Main references

1. The slides are self contained
2. Within each slides you will find references to papers and recent research in the area as optional readings
3. Optional background readings about the topic are below each module (see below). The full list of references is at the of each slide deck

For 3, I will refer to book chapter or survey articles as background readings. These can improve your understanding of the topic but are not mandatory. In particular, I will use parts of the following books (PDFs can all be found online).

• Imbens, Guido W., and Donald B. Rubin. Causal inference in statistics, social, and biomedical sciences. Cambridge University Press, 2015. (I will be using some of the chapters of this book to introduce concepts during the first couple of weeks of the course).
• Gerber, Alan S., and Donald P. Green. ``Field experiments: Design, analysis, and interpretation.” (2012). (I will be using some of the chapters of this book to introduce concepts for the analysis and design of experiments).
• Hernán MA, Robins JM (2020). Causal Inference: What If. Boca Raton: Chapman & Hall/CRC. (I will be using some of the chapters of this book to introduce some of the concepts discussed during the first couple of weeks of the course).

I will also refer sporadically to small sections of (whose PDF are freely available online)

• Lehmann and Romano (2005). Testing Statistical Hypothesis.
• Wainwright, Martin J. High-dimensional statistics: A non-asymptotic viewpoint. Vol. 48. Cambridge university press, 2019. (I will be using some sections of this book when discussing policy learning and non-asymptotic theory. The book is a useful reference if you are interested in doing research in econometrics.)
• Hastie, Trevor, Robert Tibshirani, and Martin Wainwright. Statistical learning with sparsity: the lasso and generalizations. CRC press, 2015.
• Lattimore, Tor, and Csaba Szepesvári. Bandit algorithms. Cambridge University Press, 2020.
• Pearl, Judea. Causality. Cambridge university press, 2009. (this is not available as PDF online but I will use it only in Module 2 and SLIDES on DAGs will be self-contained)

For writing the final essay, students are strongly encouraged to read

• Varanya Chaubey: The Little Book of Research Writing

The book introduces students to research writing.

Modules

Please visit the class website for updated material/more info etc

• Module 1: Introduction to Causal Inference

  1. Concepts in causal inference
  2. Selection on observables
     • Some (optional) readings: Chapter 2 of Gerber and Green; Chapter 1, 3, 12 of Imbens and Rubin, 2015
  3. Non compliance and LATE
     • Some (optional) readings: Ch. 5, 6 of Gerber and Green; Imbens and Angrist, 1994
  4. Marginal Treatment Effect
     • Some (optional) readings: Heckman, Urzua, Vytlacil (2006); Carneiro, Heckman, Vytlacil (2010)
  • Papers for in-class discussion: Imbens and Angrist, 1994; Carneiro, Heckman, Vytlacil (2010)

• Module 2: Causal Mechanisms and DAGs

  1. Intro to DAGs
     • Some (optional) readings: Ch.6 of Hernan and Robins’ book; Chapter 2, 3 of Pearl (2009)’s book; Cinelli et al. (2022, Sociological Methods andResearch); Imbens (2020, Journal of Economic Perspective)
  2. Bad and Negative Controls
     • Some (optional) readings: Lipsitch et al., 2010 (Epidemiology)
  3. Mediators and Surrogates
     • Some (optional) readings: Ch. 10 of Gerber and Green; Athey et al. (2019, unpublished)
  4. Dynamic Treatments
     • Some (optional) readings: Ch. 19, 20 of Hernan and Robins’ book; Blackwell (2013, American Journal of Political Science); Heckman and Navarro (2007, JoE); Rambachan and Shephard (2019, unpublished)
  5. Proximal methods
     • Some (optional) readings: Griliches (1977; ECMA); Miao et al. (2018; Biometrika); Deaner, 2018 (unpublished)
  • Papers for in-class discussion: Acemoglu et al (2019); Freyaldenhoven et al. 2019; Athey et al., 2019

• Module 3: Sampling and Design based uncertainty in regression analysis

  • Readings: Imbens and Rubin, Ch. 5, 8; Abadie, Athey, Imbens, Wooldridge (2020, ECMA)

• Module 4: Hypothesis Testing

  1. Hypothesis Testing and Causal Inference
     • Some (optional) readings: Ch. 3 of Gerber and Green; Ch. 4, 6, 7 in Imbens and Rubin, 2015; Freedman, 2008 (Annals of Applied Math) and Lin, 2013 (Annals of Applied Stat)
  2. Inference with clusters and spatial dependence
     • Some (optional) readings: Canay, Romano and Shaikh, 2017 (ECMA); Ibragimov and Mueller, 2016 (REStat); Conley (1999; JoE); Abadie, Athey, Imbens, Wooldrige (2022, QJE)
  3. Hypothesis testing and decision theory
     • Some (optional) readings: Lehmann and Romano (2005, Springer), Ch. 1, 3; Storey (2003, Annals of Stat); Tetenov, 2016 (unpublished); Frankel and Kasy (2022; AEJ:Micro)
  4. Multiple hypothesis testing
     • Some (optional) readings: Lehmann and Romano (2005, Springer), Ch. 9; List, Shaikh, Xu (2019, Experimental Economics); Benjamini and Hockberg (1995, JRSS-B) Kline, Rose and Walkers (2022, QJE); Viviano, Wuthrich and Niehaus (2023, Unpublished)
  • Papers for in class discussion: Mogstad et al., 2020, Kline et al., 2022

    
    

• Module 5: Complex experiments and networks

  1. Experimental Designs under SUTVA
     • Some (optional) readings: Ch. 3,4, 12 of Gerber and Green; Bugni et al. (2018, JASA); Bai (2022,AER); Max Tabord-Meehan (2022, REStud); Li and Ding (2020, JASA); Cyntrynbaum (2023, unpublished)
  2. Experiments with spillovers
     • Some (optional) readings: Ch. 8 of Gerber and Green; Baird et al. (2018, REStat); Hudgens and Halloran (2008, JASA); Aronow and Samii (2017, Journal of Applied Stat); Breza et al. (2020, AER); Banerjee et al. (2013, Science); Viviano et al. (2023, unpublished)
  3. Pre-analysis plan and Power analysis
     • Some (optional) readings: Miguel et al. (2021, Journal of Economic Perspective); Banerjee et al. (2020, NBER working paper); Kasy and Spiess (2023, unpublished)
  4. Experiments: Practice and Theory
     • Some (optional) readings: Banerjee (2020, AER); Banerjee et al. (2020, AER)
  • Papers for in class discussion: to be assigned

  
  

• Module 6: Shrinkage and robust estimation in observational studies

  1. Regression Adjustment, Doubly robustness and Augmented IPW
     • Some (optional) readings: Ch. 18 of Hernan and Robins’ book; Ch. 3 of Stefan Wager’s lecture notes; Hahn (1998, ECMA); Farrell (215, JoE); Chernozhukov et al. (2018, Econometrics Journal)
  2. Penalized regression
     • Some (optional) readings: Ch. 2 of Wainwright (2019); Ch. 11.1, 11.2 of Hastie et al.’s book (Statistical learning with sparsity)
  3. Residual Balancing in High dimensions
     • Some (optional) readings: Imai and Ratkovic (2014, JRSS-B); Zubizzareta (2015, JASA); Athey et al. (2018, JRSS-B); Bruns et al. (2023, unpublished)
  4. MSE-optimal estimation within a class of regression functions
     • Some (optional) readings: Armstrong and Kolesar (2021, ECMA); Donoho (1994, Annals of Stat)
  5. Empirical Bayes Methods
     • Some (optional) readings: NBER Method Lecture (2022); Efron (2021; Lecture note on empirical bayes); Angrist et al. (2017, ECMA)
  • Papers for in class discussion: to be assigned

    
    

• Module 7: Effects heterogeneity and policy learning

  1. Learning effects heterogeneity
     • Some (optional) readings: Kunzel et al. (2019, PNAS); Chernozhukov et al. (2018, NBER working paper); Wager and Athey (2018, JASA); Nie and Wager (2021; Biometrika); Haushofer et al. (2022, R&R AER)
  2. Statistical Treatment choice (off-line policy learning)
     • Some (optional) readings: Manski (2004, ECMA); Kitagawa and Tetenov (2019, ECMA); Athey and Wager (2021, ECMA)
  3. Adaptive experiments (online policy learning)
     • Some (optional) readings: Ch. 4, 6, 7, 11, 18, 34 of Lattimore and Szepesvari’s book; Cesabianchi and Lugosi (1999, Annals of Stat); Adusumilli (2022, R&R ECMA)
  • Papers for in class discussion: to be assigned

    
    

• Module 8: Advances in Panel Data Methods for causal inference

  1. Advances in DiD methods
     • Some (optional) references: 2023 NBER method’s lecture; survey articles Roth et al. (2023, JoE); de Chaisemartin and d’Haultfoeuille (2023, the Econometrics Journal)
  2. Synthetic Control and Matrix Completion Methods
     • Some optional readings: Abadie’s Chamberlain seminar; Abadie et al. (2010, JASA); Abadie (2021, Journal of Economic Perspective); Arkhangeleski et al. (2021, AER); Athey et al. (2021, JASA)
  3. Panel data with heterogeneity and dynamics
     • Some (optional) readings: Bonhomme and Manresa (2015, ECMA); Marx, Tamer and Tang (2022, JPE Micro); Hekman and Navarro (2007, JoE)
  • Papers for in class discussion: to be assigned

    
    

• Final Lecture: reviews and other topics