Inference in Marginal Structural Models by Automatic Targeted Bayesian and Minimum Loss-Based Estimation

TL;DR

This paper introduces novel frequentist and Bayesian methods for estimating causal effects using Marginal Structural Models, achieving efficiency bounds and providing practical algorithms with software implementation, demonstrated on real data.

Contribution

It develops a semi-parametric efficiency bound and introduces a new targeted Bayesian estimator with an automatic differentiation algorithm for flexible MSM analysis.

Findings

Achieves the semi-parametric efficiency bound for MSM parameters.

Provides a Bayesian estimator with asymptotic Bernstein von-Mises behavior.

Demonstrates methods on real-world family planning data from Malawi.

Abstract

Two of the principle tasks of causal inference are to define and estimate the effect of a treatment on an outcome of interest. Formally, such treatment effects are defined as a possibly functional summary of the data generating distribution, and are referred to as target parameters. Estimation of the target parameter can be difficult, especially when it is high-dimensional. Marginal Structural Models (MSMs) provide a way to summarize such target parameters in terms of a lower dimensional working model. We introduce the semi-parametric efficiency bound for estimating MSM parameters in a general setting. We then present a frequentist estimator that achieves this bound based on Targeted Minimum Loss-Based Estimation. Our results are derived in a general context, and can be easily adapted to specific data structures and target parameters. We then describe a novel targeted Bayesian estimator and provide a Bernstein von-Mises type result analyzing its asymptotic behavior. We propose a universal algorithm that uses automatic differentiation to put the estimator into practice for arbitrary choice of working model. The frequentist and Bayesian estimators have been implemented in the Julia software package TargetedMSM.jl. Finally, we illustrate our proposed methods by investigating the effect of interventions on family planning behavior using data from a randomized field experiment conducted in Malawi.