A Goodness-of-Fit Test for a Class of Nonhomogeneous Poisson Processes.

摘要

This paper discusses a goodness-of-fit test for a class of nonhomogeneous Poisson processes. This class of processes has been used to describe the phenomena of reliability growth and of the occurrence of failures in a complex, repairable system. Improved tables of percentiles of the small distribution of the Cramer-von Mises statistic in which a parametric exponent is estimated are provided for use in the test. The test can also be used to test the hypothesis that a random sample comes from a cumulative distribution function which can be expressed as an unknown positive power of a specified cdf. (Author)