2017/03/09_Graciela Boente, Universidad de Buenos Aires and CONICET, Argentina

Marginal estimation under a general regression model with missing responses and covariates

2017/03/09 – 12:30 h | Graciela Boente, Universidad de Buenos Aires and CONICET, Argentina

Abstract

In this talk, we will consider a general regression model where missing data occur, in the responses and in the covariates. Our aim is to discuss the estimation of any marginal functional, such as the mean, the median or any alpha-quantile of the response variable. A missing at random condition is assumed in order to prevent from bias in the estimation of the marginal measures under a non-ignorable missing mechanism. Different approaches for the estimation of the marginal functional of interest will be presented, including inverse probability weighting, the convolution of the estimators of the distributions of the errors and of the regression function and also double robust estimators, which protects against misspecication of the regression or the missing probability models. The small sample behavior of the proposals will be illustrated through a Monte Carlo study on a partially linear model.