Skew-symmetric distributions and Fisher information: the double sin of the skew-normal

DOI10.3150/13-BEJ528zbMATH Open1302.60030arXiv1209.4177OpenAlexW1971531588MaRDI QIDQ396013FDOQ396013

Marc Hallin, Christophe Ley

Publication date: 8 August 2014

Published in: Bernoulli (Search for Journal in Brave)

Abstract: Hallin and Ley [Bernoulli 18 (2012) 747-763] investigate and fully characterize the Fisher singularity phenomenon in univariate and multivariate families of skew-symmetric distributions. This paper proposes a refined analysis of the (univariate) problem, showing that singularity can be more or less severe, inducing $n^{1/4}$ ("simple singularity"), $n^{1/6}$ ("double singularity"), or $n^{1/8}$ ("triple singularity") consistency rates for the skewness parameter. We show, however, that simple singularity (yielding $n^{1/4}$ consistency rates), if any singularity at all, is the rule, in the sense that double and triple singularities are possible for generalized skew-normal families only. We also show that higher-order singularities, leading to worse-than-$n^{1/8}$ rates, cannot occur. Depending on the degree of the singularity, our analysis also suggests a simple reparametrization that offers an alternative to the so-called centred parametrization proposed, in the particular case of skew-normal and skew-$t$ families, by Azzalini [Scand. J. Stat. 12 (1985) 171-178], Arellano-Valle and Azzalini [J. Multivariate Anal. 113 (2013) 73-90], and DiCiccio and Monti [Quaderni di Statistica 13 (2011) 1-21], respectively.

Full work available at URL: https://arxiv.org/abs/1209.4177

Recommendations

• Skew-symmetric distributions and Fisher information -- a tale of two densities
• Inference on skew-normal distribution based on Fisher information in order statistics
• A note on the Fisher information matrix for the skew-generalized-normal model
• Fisher information for the elliptically symmetric Pearson distributions
• A note on the Fisher information matrix for the flexible generalized-skew-normal model
• Do tails of symmetric distributions contain more Fisher information about the scale parameter
• Properties and inference on the skew-curved-symmetric family of distributions
• Skew-rotationally-symmetric distributions and related efficient inferential procedures
• Large sample confidence intervals for the skewness parameter of the skew-normal distribution based on Fisher's transformation
• Information matrix for generalized skew-normal distributions

zbMATH Keywords

skew-normal distributionscentred parametrizationconsistency ratessingular Fisher informationskew-symmetric distributionsskewing function

Mathematics Subject Classification ID

Probability distributions: general theory (60E05) Parametric inference (62F99)

Cites Work

• Statistical Applications of the Multivariate Skew Normal Distribution
• Title not available (Why is that?)
• Title not available (Why is that?)
• The centred parameterization and related quantities of the skew-\(t\) distribution
• Robust Likelihood Methods Based on the Skew-t and Related Distributions
• Title not available (Why is that?)
• Title not available (Why is that?)
• The Skew-normal Distribution and Related Multivariate Families*
• Inferential Aspects of the Skew Exponential Power Distribution
• Distributions Generated by Perturbation of Symmetry with Emphasis on a Multivariate Skewt-Distribution
• The skew-Cauchy distribution
• Flexible Class of Skew-Symmetric Distributions
• Skewed distributions generated by the normal kernel.
• Problems of inference for Azzalini's skewnormal distribution
• A note on the asymptotic distribution of the maximum likelihood estimator for the scalar skew-normal distribution
• Title not available (Why is that?)
• Likelihood-based inference with singular information matrix
• Some skew-symmetric models
• Skew-symmetric distributions and Fisher information -- a tale of two densities
• The centred parametrization for the multivariate skew-normal distribution
• On the singularity of multivariate skew-symmetric models
• Specification testing when score test statistics are identically zero
• Title not available (Why is that?)

Cited In (13)

• Optimal tests for elliptical symmetry: specified and unspecified location
• Discussion
• Testing for sub-models of the skew \(t\)-distribution
• NEYMAN’S C(α) TEST FOR UNOBSERVED HETEROGENEITY
• Bimodality based on the generalized skew-normal distribution
• Testing for observation-dependent regime switching in mixture autoregressive models
• Skew-rotationally-symmetric distributions and related efficient inferential procedures
• A note on the Fisher information matrix for the flexible generalized-skew-normal model
• Analysis of antibody data using skew-normal and skew-\(t\) mixture models
• Skew-symmetric distributions and Fisher information -- a tale of two densities
• Large sample confidence intervals for the skewness parameter of the skew-normal distribution based on Fisher's transformation
• Detecting the direction of a signal on high-dimensional spheres: non-null and Le Cam optimality results
• Singularity Structures and Impacts on Parameter Estimation in Finite Mixtures of Distributions

Uses Software

• sn

This page was built for publication: Skew-symmetric distributions and Fisher information: the double sin of the skew-normal

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q396013)