Omnibus test of systematic treatment effect heterogeneity (Ding, Feller, Miratrix 2018)

Omnibus test of treatment effect heterogeneity along specified dimensions X. Implemented by regressing individual treatment effect on covariates and joint test of covariate coefficients = 0

Usage

dfmTest(y, a, X)

Arguments

y

outcome vector

a

treament vector

X

covariate matrix

Value

vector aith chi-squared statistic and p-value for systematic heterogeneity

References

Ding, P., A. Feller, and L. Miratrix. (2019): “Decomposing Treatment Effect Variation,” Journal of the American Statistical Association, 114, 304–17.