:package: :game_die: R/txshift: Efficient Estimation of the Causal Effects of Stochastic Interventions, with Corrections for Outcome-Dependent Sampling
This PR fixes the computation of the target parameter estimate for the inefficent implementation. It basically makes tmle_eif look more like tmle_eif_ipcw in how it computes the parameter estimate and EIF. By-hand simulations show that this seems to fix the wonky estimates seen in earlier simulations.
nhejazi
commented
6 years ago
This PR fixes the computation of the target parameter estimate for the inefficent implementation. It basically makes
tmle_eiflook more liketmle_eif_ipcwin how it computes the parameter estimate and EIF. By-hand simulations show that this seems to fix the wonky estimates seen in earlier simulations.