Choice of the hypothesis matrix for using the Anova-type-statistic

TL;DR

This paper investigates how different hypothesis matrices affect the Anova-type-statistic (ATS) in multivariate hypothesis testing, proposing methods to reduce computation time while maintaining consistent test decisions.

Contribution

It introduces a way to construct minimal companion matrices for ATS that test the same hypotheses and produce identical decisions, optimizing computational efficiency.

Findings

Constructed minimal companion matrices for ATS

Achieved consistent test decisions with reduced matrix size

Demonstrated computational savings through simulations

Abstract

Initially developed in Brunner et al. (1997), the Anova-type-statistic (ATS) is one of the most used quadratic forms for testing multivariate hypotheses for a variety of different parameter vectors $\boldsymbol{\theta}\in\mathbb{R}^d$. Such tests can be based on several versions of ATS and in most settings, they are preferable over those based on other quadratic forms, as for example the Wald-type-statistic (WTS). However, the same null hypothesis $\boldsymbol{H}\boldsymbol{\theta}=\boldsymbol{y}$ can be expressed by a multitude of hypothesis matrices $\boldsymbol{H}\in\mathbb{R}^{m\times d}$ and corresponding vectors $\boldsymbol{y}\in\mathbb{R}^m$, which leads to different values of the test statistic, as it can be seen in simple examples. Since this can entail distinct test decisions, it remains to investigate under which conditions tests using different hypothesis matrices coincide. Here, the dimensions of the different hypothesis matrices can be substantially different, which has exceptional potential to save computation effort. In this manuscript, we show that for the Anova-type-statistic and some versions thereof, it is possible for each hypothesis $\boldsymbol{H}\boldsymbol{\theta}=\boldsymbol{y}$ to construct a companion matrix $\boldsymbol{L}$ with a minimal number of rows, which not only tests the same hypothesis but also always yields the same test decisions. This allows a substantial reduction of computation time, which is investigated in several conducted simulations.