Comparing Survival Curves Using Contingency Tables

Date of Award


Degree Name

Doctor of Philosophy



First Advisor

Joshua Naranjo, Ph.D.

Second Advisor

Hyun Bin Kang, Ph.D.

Third Advisor

Duy Ngo, Ph.D.

Fourth Advisor

Mathew Rosales, Ph.D.


Chi-square test, compare k survival curves, compare two survival curves under non-pH models, contingency tables, test of homogeneity in survival analysis


Under proportional hazard assumption (PH) in survival analysis, log-rank test (Mantel, 1966) is optimal. When PH assumption is violated, Wilcoxon test (Gehan, 1965), weighted log-rank test (Fleming, Harrington, & O'Sullivan, 1987), Max- Combo test (Karrison, 2016), and restricted mean survival time (Royston & Par- mar, 2013) are alternative methods.

In this study, we propose a chi-square test of homogeneity for comparing survival curves. We propose an adaptive procedure for choosing between the number of classes in the contingency table. We also discuss an adaptive procedure based on Simes adjustment for multiple testing.

Access Setting

Dissertation-Abstract Only

Restricted to Campus until


This document is currently not available here.