do Rio do Janeiro and the 2009 Latin American Meetings of the Econometric Society. This is a heavily revised and extended version of NBER Working Paper w13981 titled "Inverse probability tilting and missing data problems". Previous versions of this paper also circulated under the title "A new method of estimating moment condition models with missing data when selection is on observables." Material in Section 4 of the initial NBER paper is not included in this version of the paper, but may be found in the companion paper "Efficient estimation of data combination problems by the method of auxiliary-to-study tilting" (NBER Working Paper w16928). All the usual disclaimers apply. The views expressed herein are those of the author(s) and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peerreviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.
for detailed comments on earlier drafts. This draft has benefited from comments by the co-editor, associate editor and three anonymous referees. We thank Jing Qin and Biao Zhang for assistance in replicating the Monte Carlo designs in Qin and Zhang (2008). We also acknowledge feedback and suggestions from participants in seminars at the
We propose a new inverse probability weighting (IPW) estimator for moment condition models with missing data. Our estimator is easy to implement and compares favorably with existing IPW estimators, including augmented inverse probability weighting (AIPW) estimators, in terms of efficiency, robustness, and higher order bias. We illustrate our method with a study of the relationship between early Black-White differences in cognitive achievement and subsequent differences in adult earnings. In our dataset the early childhood achievement measure, the main regressor of interest, is missing for many units.
for detailed comments on earlier drafts. This draft has benefited from comments by the co-editor, associate editor and three anonymous referees. We thank Jing Qin and Biao Zhang for assistance in replicating the Monte Carlo designs in Qin and Zhang (2008). We also acknowledge feedback and suggestions from participants in seminars at the
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.