Asymptotic normality for the Kφ-divergence goodness-of-fit tests

摘要

In this paper for a wide class of goodness-of-fit statistics based Kφ-divergences, the asymptotic normality is established under the assumption n/mn→a∈(0,∞), where n denotes sample size and mn the number of cells. This result is extended to contiguous alternatives to study asymptotic efficiency.