Confounding control in a nonexperimental study of STAR*D data: logistic regression balanced covariates better than boosted CART

Conclusions: Comparing multiple PS estimates is a pragmatic way to optimize balance. Logistic regression remains valuable for this purpose. Simulation studies are needed to compare PS models under varying conditions. Such studies should consider more flexible estimation methods, such as logistic models with automated selection of interactions or hybrid models using main effects logistic regression instead of a constant log-odds as the initial model for BCART.
Source: Annals of Epidemiology - Category: Epidemiology Authors: Source Type: research