A weighted Harrell–Davis distance test with applications to censored data. Issue 10 (19th May 2017)

• Record Type: Journal Article
• Authors: Wang, Dongliang; Hutson, Alan D.
• Is Part Of: Communications in statistics. Volume 46:Issue 10(2017)
• Page Start: 5022
• Page End: 5034
• Publication Date: 2017-05-19
• Subjects: Censored data -- Distance test -- Expected shortfall -- Log-rank test -- Quantile function.; Primary: 62G10 -- Secondary: 62G30; Mathematical statistics -- Periodicals; Mathematics; Statistics; 519.2
• DOI: 10.1080/03610926.2015.1096396
• Languages: English
• ISSNs: 0361-0926

Abstract

ABSTRACT: Consider the standard treatment-control model with a time-to-event endpoint. We propose a novel interpretable test statistic from a quantile function point of view. The large sample consistency of our estimator is proven for fixed bandwidth values theoretically and validated empirically. A Monte Carlo simulation study also shows that given small sample sizes, utilization of a tuning parameter through the application of a smooth quantile function estimator shows an improvement in efficiency in terms of the MSE when compared to direct application of classic Kaplan–Meier survival function estimator. The procedure is finally illustrated via an application to epithelial ovarian cancer data.