A Tutorial on Propensity Score Estimation for Multiple Treatments Using Generalized Boosted Models

By Daniel F McCaffrey, B. Griffin, D. Almirall, M. Slaughter, R. Ramchand, Lane Burgette. Statistics in Medicine, 2013

1,347 citations recorded by the source corpus.

The use of propensity scores to control for pretreatment imbalances on observed variables in non‐randomized or observational studies examining the causal effects of treatments or interventions has become widespread over the past decade. For settings with two conditions of interest such as a treatment and a control, inverse probability of treatment weighted estimation with propensity scores estimated via boosted models has been shown in simulation studies to yield causal effect estimates with desirable properties. There are tools (e.g., the twang package in R) and guidance for implementing this method with two treatments. However, there is not such guidance for analyses of three or more treatments. The goals of this paper are twofold: (1) to provide step‐by‐step guidance for researchers who want to implement propensity score weighting for multiple treatments and (2) to propose the use of generalized boosted models (GBM) for estimation of the necessary propensity score weights. We define the causal quantities that may be of interest to studies of multiple treatments and derive weighted estimators of those quantities. We present a detailed plan for using GBM to estimate propensity scores and using those scores to estimate weights and causal effects. We also provide tools for assessing balance and overlap of pretreatment variables among treatment groups in the context of multiple treatments. A case study examining the effects of three treatment programs for adolescent substance abuse demonstrates the methods. Copyright © 2013 John Wiley & Sons, Ltd.

Subfields: Education, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/fdad08d4ee7ffd2980ee534f4e6c630b62d1f4e3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Education

David M. Blei

In Statistics and Probability

Citations in the corpus

An introduction to inverse probability of treatment weighting in observational research (2021, background): “The calculation of propensity scores is not only limited to dichotomous variables, but can readily be extended to continuous or multinominal exposures [11, 12], as well as to settings involving multilevel data or competing risks [12, 13].”
Bayesian regression tree models for causal inference: regularization, confounding, and heterogeneous effects (2017, neutral): “…(TMLE) (van der Laan, 2010a,b), double machine learning (Chernozhukov et al., 2016), and generalized boosting (McCaffrey, Ridgeway and Morral, 2004; McCaffrey et al., 2013) all focus on estimation of average treatment effects, whereas our focus is on individual (heterogeneous, subgroup) treatment…”
Estimation of causal effects with multiple treatments: a review and new ideas (2017, methodology): “As another option, McCaffrey et al. (2013) used generalized boosted models to independently estimate P (Ii(t)|X), where Ii(t) = {1 if Ti = t, 0 otherwise}.”
The Blessings of Multiple Causes (2018, background): “Lopez and Gutman (2017), McCaffrey et al. (2013), Zanutto, Lu, and Hornik (2005), Rassen et al. (2011), Lechner (2001), and Feng et al. (2012) extend matching, subclassification, and weighting to multiple treatments, always assuming no unobserved confounders.”
Bayesian Regression Tree Models for Causal Inference: Regularization, Confounding, and Heterogeneous Effects (with Discussion) (2020, neutral): “…(supervised learning) methods in the service of estimation and inference for average treatment effects (ATEs) include targeted maximum likelihood estimation (TMLE) (van der Laan, 2010a,b), double machine learning (Chernozhukov et al., 2016), and generalized boosting (McCaffrey et al., 2004, 2013).”