B1460
Title: Robust inference for subgroup analysis with general transformation models
Authors: Wenxin Liu - The Chinese University of Hong Kong (Hong Kong) [presenting]
Abstract: A crucial step in developing personalized treatment strategies is to identify the subgroups of patients of a heterogeneous population. We consider a general class of heterogeneous transformation models for subgroup identification, under which an unknown monotonic transformation of the response is linearly related to the covariates via subject-specific regression coefficients with unknown error distribution and unknown {\it priori} grouping information. This class of models is broad enough to cover many popular models, including some novel heterogeneous linear models. We propose a robust method based on the maximum rank correlation and a concave fusion to automatically identify the subgroup structure and estimate the subgroup-specific treatment effects simultaneously. We establish the theoretical properties of our proposed estimate under regularity conditions. A random weighting resampling scheme is used for variance estimation. The proposed procedure can be easily extended to handle censored data.Numerical studies including simulations and a real data analysis demonstrate that the proposed method performs reasonably well in practical situations.
