B0716
Title: Maximin tests for symmetry of circular data based on the characteristic function
Authors: Thomas Verdebout - Universite Libre de Bruxelles (Belgium) [presenting]
Simos Meintanis - University of Athens (Greece)
Abstract: Inference for circular data is considered based on the empirical characteristic function. More precisely, we provide tests for reflective symmetry on the circle based on the imaginary part of the empirical characteristic function. We show that the proposed tests enjoy many attractive features. In particular, we obtain that they are locally and asymptotically maximin in the Le Cam sense under sine-skewed alternatives in the specified mean direction case. For the unspecified mean direction case, we provide corrected versions of the original tests that keep very nice asymptotic power properties. Results are illustrated on a well-known dataset and checked via Monte-Carlo simulations.
Journal