Carolina Caetano

I am an Assistant Professor of Economics at the University of Georgia. My research interests are in Microeconometrics and Health.


Address: C401 Amos Hall, 610 S. Lumpkin St., Athens, GA, 30602


Identifying multiple marginal effects with a single instrument, with Juan Carlos Escanciano, forthcoming at Econometric Theory. Paper, informal summary, code, slides.

A discontinuity test for identification in triangular nonseparable models, with Christoph Rothe and Nese Yildiz, Journal of Econometrics, v. 193(1), 2016. Paper, informal summary, slides.

A test of exogeneity without instrumental variables in models with bunching, Econometrica, v. 83(4), 2015. Paper, online appendix (details about the estimator and the simulations), code, replication files, informal summary, slides.

Working Papers

Should Children Do More Enrichment Activities? Leveraging Bunching to Correct for Endogeneity, with Gregorio Caetano and Eric Nielsen. Paper, slides. Under review.

Regression Discontinuity Design with Multivalued Treatments, with Gregorio Caetano and Juan Carlos Escanciano. Paper, slides. Under review.

Correcting Endogeneity Bias in Models with Bunching, with Gregorio Caetano and Eric Nielsen. Paper, slides.

Dummy Discontinuity Test of Exogeneity, with Gregorio Caetano, Eric Nielsen and Hao Teng. Coming soon.

The Effect of Maternal Labor Supply on Children: Evidence from Bunching, with Gregorio Caetano, Eric Nielsen and Viviane Sanfelice. Coming Soon.

Smoking Across Pregnancy and Birth Outcomes, with Kenneth Chay and Umair Khalil. Coming at some point, I promise.

Teaching Notes and Slides

Graduate Econometrics: causal identification and estimation

1. Fundamentals of Causal Inference, slides
2. Review of OLS, slides
3. Review of the Linear Model, slides
4. Problems with the Linear Model, slides
5. Instrumental Variable Methods, slides
6. Potential Outcomes Model, slides
7. Regression Discontinuity Design, slides
8. Panel Data Models, slides
9. Statistical Learning

Nonparametric Econometrics

1. Kernel Based Methods
2. Bandwidth choice, Inference and oral Polynomial
3. Sieve Based Methods, Splines
4. Wavelets, Choice of Series Length, and Inference
5. Semi-parametric Models: Partially Linear, Additive, Varying Coefficients, Single Index
6. Examples: Nonparametric IV, Inverse Prob. Weighted Propensity Score, RDD