Goodness-of-fit tests using a novel estimator of Phi-divergence

This paper introduces a novel estimator for the Phi-divergence criterion and provides a general goodness-of-fit test. Desirable properties, such as consistency and invariance of the proposed test, are established. The paper also investigates some special cases of tests for Weibull, log-normal, and exponential distributions. By conducting a simulation study, the power values of the proposed test are compared with those of some well-known competing tests under various alternatives. Finally, two examples are provided to illustrate the proposed procedure.