A1420
Title: Distance covariance for random fields
Authors: Muneya Matsui - Nanzan University (Japan) [presenting]
Abstract: An independence test is studied based on distance correlation for random fields (X,Y). The situations are considered when (X,Y) is observed on a lattice with equidistant grid sizes and when (X,Y) is observed at random locations. The asymptotic theory is provided for the sample distance correlation in both situations, and bootstrap consistency is shown. The latter fact allows one to build a test for independence of X and Y based on the considered discretization of these fields. The performance of the bootstrap test is illustrated in a simulation study involving fractional Brownian and infinite variance stable fields. The independence test is applied to Japanese meteorological data, which are observed over the entire area of Japan.
