Title: Goodness-of-fit testing for normal mixtures
Authors: Dimitrios Bagkavos - University of Ioannina (Greece) [presenting]
Abstract: A novel goodness-of-fit test is introduced for the case where the hypothesized distribution is a mixture of normal distributions. The theoretical results contributed include analytic quantification of the test statistic asymptotic distribution under both the null and the alternative hypothesis and closed-form expressions for its power under Pitman alternatives, assuming a parametrically estimated underlying model. Further, the Edgeworth expansion of the proposed test's size and power functions are derived and employed in developing a bandwidth selector, designed to optimize power, subject to keeping the size constant. Numerical examples illustrate the performance of the proposed test in practice.