Assessment of Optimal Selected Prognostic Factors
Science Direct Working Paper No S1574-0358(04)70335-4
23 Pages Posted: 20 Mar 2018
Date Written: October 2002
Abstract
The identification and assessment of prognostic factors is one of the major tasks in clinical research. The assessment of one single prognostic factor can be done by recently established methods for using optimal cutpoints. Here, we suggest a method to consider an optimal selected prognostic factor from a set of prognostic factors of interest. This can be viewed as a variable selection method and is the underlying decision problem at each node of various tree building algorithms.We propose to use maximally selected statistics where the selection is defined over the set of prognostic factors and over all outpoints in each prognostic factor. The maximum of multivariate normally distributed random variables is the approximate null distribution for maximally selected rank statistics. We demonstrate that it is feasible to compute the approximate null distribution. Data of the German Breast Cancer Study Group and of a small study on patients with diffuse large B-cell lymphoma illustrate our method. Moreover, we discuss the wide range of possible applications of our suggestion.
Keywords: prognostic factor, clinical research, statistical computing, variable selection, maximally selected tests, -value adjusted regression trees
Suggested Citation: Suggested Citation