Model Averaging for Generalized Linear Model with Covariates that are Missing completely at Random

TL;DR

This paper introduces a model averaging approach for estimating generalized linear models with completely missing covariates, demonstrating its asymptotic optimality and superior performance through simulations.

Contribution

The paper proposes a novel model averaging estimator for GLMs with missing covariates that is proven to be asymptotically optimal.

Findings

The proposed method outperforms existing alternatives in simulations.

The estimator is asymptotically optimal under certain assumptions.

Abstract

In this paper, we consider the estimation of generalized linear models with covariates that are missing completely at random. We propose a model averaging estimation method and prove that the corresponding model averaging estimator is asymptotically optimal under certain assumptions. Simulaiton results illustrate that this method has better performance than other alternatives under most situations.