Estimation of propensity scores using generalized additive models.

Journal Article (Journal Article)

Propensity score matching is often used in observational studies to create treatment and control groups with similar distributions of observed covariates. Typically, propensity scores are estimated using logistic regressions that assume linearity between the logistic link and the predictors. We evaluate the use of generalized additive models (GAMs) for estimating propensity scores. We compare logistic regressions and GAMs in terms of balancing covariates using simulation studies with artificial and genuine data. We find that, when the distributions of covariates in the treatment and control groups overlap sufficiently, using GAMs can improve overall covariate balance, especially for higher-order moments of distributions. When the distributions in the two groups overlap insufficiently, GAM more clearly reveals this fact than logistic regression does. We also demonstrate via simulation that matching with GAMs can result in larger reductions in bias when estimating treatment effects than matching with logistic regression.

Full Text

Duke Authors

Cited Authors

  • Woo, M-J; Reiter, JP; Karr, AF

Published Date

  • August 2008

Published In

Volume / Issue

  • 27 / 19

Start / End Page

  • 3805 - 3816

PubMed ID

  • 18366144

Electronic International Standard Serial Number (EISSN)

  • 1097-0258

International Standard Serial Number (ISSN)

  • 0277-6715

Digital Object Identifier (DOI)

  • 10.1002/sim.3278


  • eng