Testing the Validity of the Exponential Model Based on Type II Censored Data Using Transformed Sample Data

Abstract

In this article, we propose a new approach to testing exponentiality for Type-II censored samples. Our method converts the original observations into transformed data which follow from a special distribution. It is shown that the proposed tests do not depend on the scale parameter of the exponential distribution and hence the tests are exact. Power performance of the proposed tests is compared with the existing tests against various alternatives. The results are impressive and the proposed tests have higher power than the competing tests. Finally, two real data from reliability literature as illustrate examples are presented.