eummd

CRANeummdMaRDI QIDQ5981612FDOQ5981612

Efficient Univariate Maximum Mean Discrepancy

Dean A. Bodenham

Last update: 2 October 2023

Copyright license: GNU General Public License, version 3.0, GNU General Public License, version 2.0

Software version identifier: 0.0.7, 0.1.1, 0.1.4

Description

Computes maximum mean discrepancy two-sample test for univariate data using the Laplacian kernel, as described in Bodenham and Kawahara (2023) <doi:10.1007/s11222-023-10271-x>. The p-value is computed using permutations. Also includes implementation for computing the robust median difference statistic 'Q_n' from Croux and Rousseeuw (1992) <doi:10.1007/978-3-662-26811-7_58> based on Johnson and Mizoguchi (1978) <doi:10.1137/0207013>.

Cites work

• Selecting the Kth Element in $X + Y$ and $X_1 + X_2 + \cdots + X_m $
• euMMD: efficiently computing the MMD two-sample test statistic for univariate data
• Time-Efficient Algorithms for Two Highly Robust Estimators of Scale

This page was built for software: eummd