Exponentiality test based on Renyi distance between equilibrium distributions |
| |
Authors: | M Sadeghpour A Habibirad |
| |
Institution: | Department of Statistics, School of Mathematical Sciences, Ferdowsi University of Mashhad, Mashhad, Iran |
| |
Abstract: | ABSTRACTOn the basis of Csiszar's φ-divergence discrimination information, we propose a measure of discrepancy between equilibriums associated with two distributions. Proving that a distribution can be characterized by associated equilibrium distribution, a Renyi distance of the equilibrium distributions is constructed that made us to propose an EDF-based goodness-of-fit test for exponential distribution. For comparing the performance of the proposed test, some well-known EDF-based tests and some entropy-based tests are considered. Based on the simulation results, the proposed test has better powers than those of competing entropy-based tests for the alternatives with decreasing hazard rate function. The use of the proposed test is evaluated in an illustrative example. |
| |
Keywords: | Csiszar's divergence Cumulative residual entropy Equilibrium distribution Exponential distribution Goodness-of-fit test Kullback–Leibler divergence |
|
|