Estimating equations under IPW imputation of missing data

Abstract

The inverse probability weighted (IPW) imputation method is first applied to compensate for non-response. And then, the empirical likelihood (EL) inference is made for estimation equation parameters. It is a nice result to be obtained in this paper that the limiting distributions of the EL statistics are χ2-type distributions under the IPW imputation. Compared with the usually-used methods, the proposed method is easier to complement and performs more efficient.