Distributional (Single) Index Models

Abstract

A Distributional (Single) Index Model (DIM) is a semi-parametric model for dis-tributional regression, that is, estimation of conditional distributions given covariates.The method is a combination of classical single index models for the estimation of theconditional mean of a response given covariates, and isotonic distributional regression.The model for the index is parametric, whereas the conditional distributions are esti-mated non-parametrically under a stochastic ordering constraint. We show consistencyof our estimators and apply them to a highly challenging data set on the length of stay(LoS) of patients in intensive care units. We use the model to provide skillful andcalibrated probabilistic predictions for the LoS of individual patients, that outperformthe available methods in the literature.