A1967
Title: An RP-MMD framework for multivariate goodness-of-fit testing in location-scale families
Authors: Xiaoling Peng - Beijing Normal-Hong Kong Baptist University (China) [presenting]
Abstract: With the increasing prevalence of high-dimensional data, developing goodness-of-fit tests that are both statistically powerful and computationally efficient remains a fundamental challenge. A scalable multivariate goodness-of-fit testing framework, termed the RP-MMD test, is proposed for a broad class of location-scale distributions. Extensive simulation studies demonstrate that the proposed method achieves competitive or superior power compared with existing tests, including the Shapiro--Wilk test, across a wide range of scenarios. Applications to real data further illustrate the practical effectiveness of the approach.
