tmle3shift: Targeted Learning of the Causal Effects of Stochastic Shift Interventions

tmle3shift is an adapter/extension R package in the tlverse ecosystem that exposes support for the estimation of a target parameter defined as the mean counterfactual outcome under a shift of the natural value of a continuous-valued intervention. The package provides tools for targeted maximum likelihood estimation (TMLE) of the population-level causal effects of stochastic treatment regimes characterized as modified treatment policies and related nonparametric variable importance analyses.