car_adj

Bugni, Canay, and Shaikh (2017) show that when treatment status is assigned in a covariate-adaptive randomization scheme (i.e., stratifying according to baseline covariates and then assigning treatment status to achieve within-stratum "balance"), traditional heteroskedasticity-consistent standard errors may be biased. -car_sat- calculates a standard error adjustment for this bias in a "fully saturated" regression; that is, a linear regression of the outcome variable on all interactions between indicators for each of the treatments and indicators for each of the strata. -car_sfe- calculates a standard error adjustment for this bias in a "strata fixed-effects" regression; that is, a linear regression of the outcome variable on indicators for each of the treatments and indicators for each of the strata.