Abstract
This article presents method of constructing tests for the hypothesis on shape parameter 𝛽 of the family of lifetime distributions under progressive censoring. We derive three test statistics 𝑄(𝑠𝑑), F and 𝑄1. The performance of these statistics, in terms of power, is studied through simulation for different values of shape parameter of Gompertz distribution, Burr-XII distribution, bathtub distribution, and Weibull distribution. For these distributions, it is concluded that either a test 𝑄(𝑠𝑑), or test 𝑄1 performs better than test F in case of large samples whereas test F performs better than other tests 𝑄(𝑠𝑑)and 𝑄1 in case of small samples when the shape parameter is not equal to the value under null hypothesis.
Author: Ashok Shanubhogue and Rajendra G. Desai
Received on: November, 2018
Accepted on: October, 2019