chizhangucb/tmleCommunity: Targeted Maximum Likelihood Estimation for Hierarchical Data

Targeted minimum loss-based estimation (TMLE) of the average causal effect of community-based intervention(s) at a single time point on an individual-based outcome of interest. It provides three approaches to analyze hierarchical data: community-level TMLE, inidividual-level TMLE and stratified TMLE. Implementations of the inverse-probability-of- treatment-weighting (IPTW) and the G-computation formula (GCOMP) are also available for each approach. The package supports multivariate arbitrary interventions (deterministic or stochastic) with a binary or continuous outcome. The tmleCommunity() function calculates the marginal treatment effect among independent community units (or i.i.d individual units if no hierarchical structure) using TMLE. Besides, it allows user-specified data-adaptive machine learning algorithms through SuperLearner and h2oEnsemble packages. The input dataset should be made up of rows of community-specific and individual-specific observations, with each row i (in community j) containing random variables (W_{i,j}, E_j, A_j, Y_{i,j}), where E_j represents a vector of community j's environmental baseline covariates, W_{i,j}represents a vector of individual i's individual-level baseline covariates, A_j is the exposure(s) (can be univariate or multivariate, can be binary, categorical or continuous) assigned or naturally occurred in community j and Y_{i,j} is i's outcome (either binary or continuous). More details can be found in '?tmleCommunity-package' and '?tmleCommunity'.

install.packages("remotes")
remotes::install_github("chizhangucb/tmleCommunity")