Login with your Reed credentials to view all events.

Targeted Function Balancing -
This talk introduces Targeted Function Balancing (TFB), a covariate balancing framework for estimating the causal effect of a binary intervention on an outcome. TFB first regresses the outcome on observed covariates, and then selects weights that balance functions (of the covariates) that are probabilistically near the resulting regression function. Notably, TFB demonstrates that intentionally leaving imbalance in some covariates can increase efficiency without introducing bias, challenging traditions that warn against imbalance in any variable. Further, TFB is entirely defined by the choice of regression estimator and an estimator for the resulting function’s variance, turning the problem of how best to balance the covariates into how best to model the outcome, and inviting the application of machine learning models.

Event Details

0 people are interested in this event