Imperial College London
Portrait Picture

DrDeanBodenham

Faculty of Natural Sciences, Department of Mathematics

Lecturer in Statistics
 
 
 
//

Contact

 

dean.bodenham

 
 
//

Location

 

531Huxley BuildingSouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@article{Bodenham:2023:10.1007/s11222-023-10271-x,
author = {Bodenham, DA and Kawahara, Y},
doi = {10.1007/s11222-023-10271-x},
journal = {Statistics and Computing},
title = {euMMD: efficiently computing the MMD two-sample test statistic for univariate data},
url = {http://dx.doi.org/10.1007/s11222-023-10271-x},
volume = {33},
year = {2023}
}
Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - The maximum mean discrepancy (MMD) test is a nonparametric kernelised two-sample test that, when using a characteristic kernel, can detect any distributional change between two samples. However, when the total number of d-dimensional observations is n, direct computation of the test statistic is O(dn2). While approximations with lower computational complexity are known, more efficient methods for computing the exact test statistic are unknown. This paper provides an exact method for computing the MMD test statistic for the univariate case in O(nlogn) using the Laplacian kernel. Furthermore, this exact method is extended to an approximate method for d-dimensional real-valued data also with complexity log-linear in the number of observations. Experiments show that this approximate method can have good statistical performance when compared to the exact test, particularly in cases where d>n.
AU  - Bodenham,DA
AU  - Kawahara,Y
DO  - 10.1007/s11222-023-10271-x
PY  - 2023///
SN  - 0960-3174
TI  - euMMD: efficiently computing the MMD two-sample test statistic for univariate data
T2  - Statistics and Computing
UR  - http://dx.doi.org/10.1007/s11222-023-10271-x
UR  - http://hdl.handle.net/10044/1/107066
VL  - 33
ER  -
Download