Testing the equality of multivariate means when p>n by combining the Hoteling and Simes tests

12/22/2019
by   Tzviel Frostig, et al.
0

We propose a method of testing the shift between mean vectors of two multivariate Gaussian random variables in a high-dimensional setting incorporating the possible dependency and allowing p > n. This method is a combination of two well-known tests: the Hotelling test and the Simes test. The tests are integrated by sampling several dimensions at each iteration, testing each using the Hotelling test, and combining their results using the Simes test. We prove that this procedure is valid asymptotically. This procedure can be extended to handle non-equal covariance matrices by plugging in the appropriate extension of the Hotelling test. Using a simulation study, we show that the proposed test is advantageous over state-of-the-art tests in many scenarios and robust to violation of the Gaussian assumption.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
11/16/2020

Covariance matrix testing in high dimension using random projections

Estimation and hypothesis tests for the covariance matrix in high dimens...
research
08/17/2021

Stable combination tests

This paper proposes a stable combination test, which is a natural extens...
research
12/28/2020

Rao's Score Tests on Correlation Matrices

Even though the Rao's score tests are classical tests, such as the likel...
research
05/15/2018

Visualizing Tests for Equality of Covariance Matrices

This paper explores a variety of topics related to the question of testi...
research
10/08/2021

Joint Normality Test Via Two-Dimensional Projection

Extensive literature exists on how to test for normality, especially for...
research
11/01/2020

Data Analysis for Proficiency Testing

Proficiency Testing (PT) determines the performance of individual labora...
research
02/28/2020

Statistical applications of Random matrix theory: comparison of two populations II

This paper investigates a statistical procedure for testing the equality...

Please sign up or login with your details

Forgot password? Click here to reset