More good news on the HKM test for multivariate reflected symmetry about an unknown centre (Q2183765)

The authors re-examine testing multivariate reflected symmetry about an unspecified point. Let \(X\), \(X_1\), \(X_2,\dots\) be a sequence of independent and identically distributed \(d\)-dimensional random vectors. They investigate the problem of testing the hypothesis \[ H_0 : X-\mu \stackrel{\mathcal{D}}{=} \mu - X \] for some unknown \(\mu \in \mathbb{R}^{d}\), against general alternatives. The notation \(\stackrel{\mathcal{D}}{=}\) denotes equality in distribution. The authors consider the test statistic \( T_{n,a}=T_{n,a}(X_1,\dots,X_n)\) of \textit{N. Henze} et al. [J. Multivariate Anal. 87, No. 2, 275--297 (2003; Zbl 1040.62047)], where \(a>0\) is some fixed parameter. From the authors' abstract: We identify a measure of deviation \(\Delta_a\)(say) from symmetry associated with the test statistic \(T_{n,a}\) (say), and we obtain the limit normal distribution of \(T_{n,a}\) as \(n \to \infty\) under a fixed alternative to symmetry. Since a consistent estimator of the variance of this limit normal distribution is available, we obtain an asymptotic confidence interval for \(\Delta_a\). The test, when applied to a classical data set, strongly rejects the hypothesis of reflected symmetry, although other tests even do not object against the much stronger hypothesis of elliptical symmetry.