I am a Ph.D. candidate in the Department of Economics at the University of California, Santa Barbara. My research interests include Econometrics and Applied Econometrics. Before joining the program, I was a Research Fellow in the Research Department at the Inter-American Development Bank in Washington, D.C.
At What Level Should One Cluster Standard Errors in Paired Experiments, and in Stratified Experiments with Small Strata? (joint with Clément de Chaisemartin) Available on here. (Submitted)
Abstract: In paired experiments, units are matched into pairs, and one unit of each pair is randomly assigned to treatment. To estimate the treatment effect, researchers often regress their outcome on a treatment indicator and pair fixed effects, clustering standard errors at the unit-of-randomization level. We show that the variance estimator in this regression may be severely downward biased: under constant treatment effect, its expectation equals 1/2 of the true variance. Instead, we show that researchers should cluster their standard errors at the pair level. Using simulations, we show that those results extend to stratified experiments with few units per strata.