Abstract
This article introduces a novel, Bayesian, semi-parametric approach to inference for both elliptical and normal/independent distributions. The location and scale parameters are modelled parametrically and a suitable transformation of the modular variable is modelled using Dirichlet process mixtures. A feature of our approach is that the partial lack of identifiability inherent in both elliptical and normal/independent distributions can be accounted for by incorporating a restriction on the diagonal elements of the scale matrix. Posterior computation is carried out using a Markov chain Monte Carlo algorithm.A novel technique for model selection, based on an approximation of the deviation information criterion, is introduced. As shown by a numerical study based on simulation, the approach can be used to discriminate between elliptical, and normal/independent distributions. Finally, our methodology is illustrated with both simulated and real data.
Funding Information
  • Junta de Andalucía, Spain (FQM-329, US-1381178, P18-FR-2369)
  • Ministerio de Economía, Industria y Competitividad, Spain (PID2019-110886RB-I00)
  • Universidad de Cádiz, Spain (PR2019-029)
  • AEI  (TED2021-130216A-I00)
  • Agencia Estatal de Investigación of Spain (2021/00384/001)
  • AEI 
  • Agencia Estatal de Investigación of Spain
  • Catedra de Excelencia de la Comunidad de Madrid
  • Universidad Carlos III de Madrid
  • ICMAT-CSIC

This publication has 47 references indexed in Scilit: