Estimation of a semiparametric natural direct effect model incorporating baseline covariates

Establishing cause-effect relationships is a standard goal of empirical science. Once the existence of a causal relationship is established, the precise causal mechanism involved becomes a topic of interest. A particularly popular type of mechanism analysis concerns questions of mediation, i.e., to what extent an effect is direct, and to what extent it is mediated by a third variable. A semiparametric theory has recently been proposed that allows multiply robust estimation of direct and mediated marginal effect functionals in observational studies (Tchetgen Tchetgen & Shpitser, 2012). In this paper we extend the theory to handle parametric models of natural direct and indirect effects within levels of pre-exposure variables with an identity or log link function, where the model for the observed data likelihood is otherwise unrestricted. We show that estimation is generally infeasible in such a model because of the curse of dimensionality associated with the required estimation of auxiliary conditional densities or expectations, given high-dimensional covariates. Thus, we consider multiply robust estimation and propose a more general model which assumes that a subset, but not the entirety, of several working models holds.