A Generalized Order-restricted Inference Methodology for Selecting and Clustering Genes

摘要

There are many methods for selecting and clustering genes according to their time-course or dose-response profiles. These methods all necessitate the assumption of a constant variance through time or among dosages. This homoscedasticity assumption is, however, seldom satisfied in practice. In this paper, via the application of Shi’s (1994,1998) algorithms and a modified bootstrap procedure, we proposed a generalized order-restricted inference methodology for the same task without the homoscedasticity restriction. Simulation results show that our procedure can control the false positive rate and have some good qualities.Key words: level probability / 2 E2 test / bootstrap sampling / PAVA algorithm