On bias correction in small area estimation: An M-quantile approach

In this paper we propose two bias correction approaches in order to reduce the prediction bias of the robust M-quantile predictors in small area estimation in the presence of representative outliers. A bootstrap procedure is considered for the estimation of the mean squared error. A Monte-Carlo simulation study is conducted. Results confirm that our approaches improve the efficiency and reduce the predic- tion bias of M-quantile predictors when the population contains units that may be influential if selected in the sample.