Testing Whether a Survival Function Is New Better than Used of a Specified Age

Document Type : Original Article

Authors

Department of Statistics & O.R King Saud University - Saudi Arabia

Abstract

A survival variable is a nonnegative random variable X with distribution function F and a survival function F=1-F. This variable is said to be new better than used of a specified age t0 if
F (x+t0) ≤ F (x) F (t0) for all x≥0 and a fixed t0. This is a large and practical class of life distributions. Its properties, applicability, and testing were discussed by Hollander, Park and Proschan (1986), (HPP). In the current investigation, it is demonstrated that a goodness of fit approach is possible to carry out this testing problem and that it results in simpler and asymptotically equivalent procedure to the HPP test.
 

Keywords