Improved double-robust estimation in missing data and causal inference models

ROTNITZKY, A.; LEI, Q.; SUED, M.; ROBINS, J.

BIOMETRIKA

OXFORD UNIV PRESS

Lugar: Oxford; Año: 2012 p. 439 - 456

0006-3444

Recently proposed double-robust estimators for a population mean from incompletedata and for a fi…nite number of counterfactual means can have much higher efficiency thanthe usual double-robust estimators under misspeci…cation of the outcome model. In thispaper we derive a new class of double-robust estimators for the parameters of regressionmodels with incomplete cross-sectional or longitudinal data, and of marginal structuralmean models for cross-sectional data with similar efficiency properties. Unlike the recentproposals, our estimators solve outcome regression estimating equations. In a simulationstudy, the new estimator shows improvements in variance relative to the standard double-robust estimator that are in agreement with those suggested by asymptotic theory.