A0769
Title: A High Dimensional Cramer-von Mises Test
Authors: Danna Zhang - University of California, San Diego (United States)
Mengyu Xu - University of Central Florida (United States) [presenting]
Abstract: We develop a Cram\'er-von Mises type test for testing distributions of high dimensional continuous data and establish an asymptotic theory for quadratic functions of high-dimensional stochastic processes. To obtain cutoff values of our tests, we introduce two different procedures to implement high-dimensional Cram\'er-von Mises test in practice: a plug-in calibration method and subsampling method. Theoretical justification and numerical studies of both approaches are provided. The method is applied to test the marginal normality of residuals from a high-dimensional vector autoregression of a macro-economic dataset.
