HETEROSKEDASTICITY-ROBUST INFERENCE IN LINEAR REGRESSION MODELS WITH MANY COVARIATES

Résumé

We consider inference in linear regression models that is robust to heteroskedasticity and the presence of many control variables. When the number of control variables increases at the same rate as the sample size the usual heteroskedasticity-robust estimators of the covariance matrix are inconsistent. Hence, tests based on these estimators are size distorted even in large samples. An alternative covariance-matrix estimator that remains consistent is presented. Unlike the estimator of Cattaneo, Jansson and Newey (2018) this estimator remains well-dened in the presence of highly-in uential observations. Simulation results and an empirical illustration are also provided.