Powerful generalized sign tests based on sign depth
Loading...
Date
2020
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The classical sign test usually provides very bad power for certain alternatives. We present
a generalization which is similarly easy to comprehend but much more powerful. It is based on
K-sign depth, shortly denoted by K-depth. These so-called K-depth tests are motivated by
simplicial regression depth, but are not restricted to regression problems. They can be applied
as soon as the true model leads to independent residuals with median equal to zero. Moreover,
general hypotheses on the unknown parameter vector can be tested. Since they depend only
on the signs of the residuals, these test statistics are outlier robust. While the 2-depth test, i.e.
the K-depth test for K = 2, is equivalent to the classical sign test, K-depth test with K ≥3
turn out to be more powerful in many applications. As we will briefly discuss, these tests are
also related to runs tests. A drawback of the K-depth test is its fairly high computational effort
when implemented naively. However, we show how this inherent computational complexity can
be reduced. In order to see why K-depth tests with K ≥ 3 are more powerful than the classical
sign test, we discuss the asymptotic behaviour of its test statistic for residual vectors with only
few sign changes, which is in particular the case for some nonfits the classical sign test cannot
reject. In contrast, we also consider residual vectors with alternating signs, representing models
that fit the data very well. Finally, we demonstrate the good power of the K-depth tests for
quadratic regression.
Description
Table of contents
Keywords
K-sign depth, quadratic regression, distribution free, outlier robust, runs test, sign test