B0406
Title: Doubly flexible estimation under label shift
Authors: Yanyuan Ma - PSU (United States) [presenting]
Abstract: In many studies, complete data are available from a population P, but the quantity of interest is often sought for a different population Q which only has partial data. The setting is considered that both outcome Y and covariate X are available from P whereas only X is available from Q, under the so-called label shift assumption. To estimate the parameter of interest in population Q via leveraging the information from population P, the following three ingredients are essential: (a) the common conditional distribution of X given Y, (b) the regression model of Y given X in population P, and (c) the density ratio of the outcome Y between the two populations. An estimation procedure is proposed that only needs some standard nonparametric regression technique to approximate the conditional expectations with respect to (a), while by no means needs an estimate or model for (b) or (c). The large sample theory is developed for the proposed estimator and its finite-sample performance is examined through simulation studies as well as an application to the MIMIC-III database.
