Semiparametric efficiency in GMM models with auxiliary data

Abstract

We study semiparametric efficiency bounds and efficient estimation of parameters defined through general moment restrictions with missing data. Identification relies on auxiliary data containing information about the distribution of the missing variables conditional on proxy variables that are observed in both the primary and the auxiliary database, when such distribution is common to the two data sets. The auxiliary sample can be independent of the primary sample, or can be a subset of it. For both cases, we derive bounds when the probability of missing data given the proxy variables is unknown, or known, or belongs to a correctly specified parametric family. We find that the conditional probability is not ancillary when the two samples are independent. For all cases, we discuss efficient semiparametric estimators. An estimator based on a conditional expectation projection is shown to require milder regularity conditions than one based on inverse probability weighting.

Published in at http://dx.doi.org/10.1214/009053607000000947 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Cited by in corpus (20)

• Robust Inference on Average Treatment Effects with Possibly More Covariates than Observations
• Deep Neural Networks for Estimation and Inference
• Empirical likelihood for estimating equations with missing values
• Estimating Treatment Effects using Multiple Surrogates: The Role of the Surrogate Score and the Surrogate Index
• Contamination Bias in Linear Regressions
• Estimating linear functionals in nonlinear regression with responses missing at random
• Finite-Sample Optimal Estimation and Inference on Average Treatment Effects Under Unconfoundedness
• Overlap in Observational Studies with High-Dimensional Covariates
• On the role of surrogates in the efficient estimation of treatment effects with limited outcome data
• Regularized Orthogonal Machine Learning for Nonlinear Semiparametric Models
• Causal Inference of General Treatment Effects using Neural Networks with A Diverging Number of Confounders
• Efficient nonparametric estimation of causal mediation effects
• Generalized Method of Moments Estimator Based On Semiparametric Quantile Regression Imputation
• Semiparametric Estimation of Long-Term Treatment Effects
• Estimation and Inference for Policy Relevant Treatment Effects
• An efficient and doubly robust empirical likelihood approach for estimating equations with missing data
• Two Sample Unconditional Quantile Effect
• Semiparametric Efficiency Gains From Parametric Restrictions on Propensity Scores
• Sharp Bounds for the Marginal Treatment Effect with Sample Selection
• GEAR: On Optimal Decision Making with Auxiliary Data