A Minimum Distance Estimation Approach to the Two-Sample Location-Scale Problem

Authors: Zhang Z.¹; Yu Q.²

Source: Lifetime Data Analysis, Volume 8, Number 3, September 2002 , pp. 289-305(17)

Publisher: Springer

Buy & download fulltext article:

Price: $47.00 plus tax (Refund Policy)

Abstract:

As reported by Kalbfleisch and Prentice (1980), the generalized Wilcoxon test fails to detect a difference between the lifetime distributions of the male and female mice died from Thymic Leukemia. This failure is a result of the test's inability to detect a distributional difference when a location shift and a scale change exist simultaneously. In this article, we propose an estimator based on the minimization of an average distance between two independent quantile processes under a location-scale model. Large sample inference on the proposed estimator, with possible right-censorship, is discussed. The mouse leukemia data are used as an example for illustration purpose.

Keywords: location-scale model; censored data; two-sample problem; quantile

Language: English

Document Type: Regular paper

Affiliations: 1: Department of Mathematics, University of North Carolina at Charlotte, NC 28223 2: Department of Mathematical Sciences, SUNY at Binghamton, NY 13902

Publication date: 2002-09-01