A stationary bootstrap test about two mean vectors

Comparison with somewhat dense differences and fewer

Sample size than dimension

Zhengbang Li 1 · Fuxiang Liu 2 · Luanjie Zeng 1 · Guoxin Zuo 1

Abstract

Two sample mean vectors comparison hypothesis testing problems often emerge in  modern biostatistics. Many tests are proposed for detecting relatively dense signals with somewhat dense nonzero components in mean vectors differences. One kind 

of these tests is based on some quadratic forms about two sample mean vectors differ- ences. Another kind of these tests is

based on some quadratic forms about studentized version of two sample mean vectors differences. In this article, we propose a boot-strap test by adopting stationary bootstrap scheme to calculate p value of a typical testwhich is based on a quadratic form about studentized version of two sample mean vectors differences. Extensive simulations are conducted to compare 

performances of the bootstrap test with other existing typical tests. We also apply the bootstrap test to a real genetic data analysis about breast cancer.

Keywords High dimens