Bayesian estimation of a bivariate copula using the Jeffreys prior

DOI10.3150/10-BEJ345zbMATH Open1318.62112arXiv0908.2372MaRDI QIDQ418233FDOQ418233

François Perron, Simon Guillotte

Publication date: 28 May 2012

Published in: Bernoulli (Search for Journal in Brave)

Abstract: A bivariate distribution with continuous margins can be uniquely decomposed via a copula and its marginal distributions. We consider the problem of estimating the copula function and adopt a Bayesian approach. On the space of copula functions, we construct a finite-dimensional approximation subspace that is parametrized by a doubly stochastic matrix. A major problem here is the selection of a prior distribution on the space of doubly stochastic matrices also known as the Birkhoff polytope. The main contributions of this paper are the derivation of a simple formula for the Jeffreys prior and showing that it is proper. It is known in the literature that for a complex problem like the one treated here, the above results are difficult to obtain. The Bayes estimator resulting from the Jeffreys prior is then evaluated numerically via Markov chain Monte Carlo methodology. A rather extensive simulation experiment is carried out. In many cases, the results favour the Bayes estimator over frequentist estimators such as the standard kernel estimator and Deheuvels' estimator in terms of mean integrated squared error.

Full work available at URL: https://arxiv.org/abs/0908.2372

Recommendations

• Bayesian nonparametric estimation of a copula
• scientific article
• A bayesian estimator for the dependence function of a bivariate extreme‐value distribution
• Bayesian nonparametric inference for a multivariate copula function
• Introduction to Bayesian Estimation and Copula Models of Dependence

zbMATH Keywords

copulaMarkov chain Monte CarlononparametricBirkhoff polytopedoubly stochastic matricesfinite mixturesobjective BayesMetropolis-within-Gibbs sampling

Mathematics Subject Classification ID

Bayesian inference (62F15) Nonparametric estimation (62G05) Characterization and structure theory for multivariate probability distributions; copulas (62H05)

Cites Work

• Title not available (Why is that?)
• Extending the rank likelihood for semiparametric copula estimation
• Statistical analysis of finite mixture distributions
• Strong approximation of copulas
• Asymptotic efficiency of the two-stage estimation method for copula-based models
• Title not available (Why is that?)
• A Primer on Copulas for Count Data
• An introduction to copulas.
• Comparison of semiparametric and parametric methods for estimating copulas
• Title not available (Why is that?)
• Title not available (Why is that?)
• THE BERNSTEIN COPULA AND ITS APPLICATIONS TO MODELING AND APPROXIMATIONS OF MULTIVARIATE DISTRIBUTIONS
• Title not available (Why is that?)
• A semiparametric estimation procedure of dependence parameters in multivariate families of distributions
• Markov chain Monte Carlo. Stochastic simulation for Bayesian inference.
• On rank correlation measures for non-continuous random variables
• Density estimation
• Copula modeling: An introduction for practitioners
• The Ehrhart polynomial of the Birkhoff polytope
• Title not available (Why is that?)
• Bayesian inference for contingency tables with given marginals
• Title not available (Why is that?)
• Results and problems in the theory of doubly-stochastic matrices

Cited In (8)

• Bayesian estimation of generalized partition of unity copulas
• Prior distributions for objective Bayesian analysis
• Bayesian nonparametric estimation of a copula
• Bayesian nonparametrics for directional statistics
• Copula modelling with penalized complexity priors: the bivariate case
• Introduction to Bayesian Estimation and Copula Models of Dependence
• Bayesian Estimation for Bivariate Gamma Processes with Copula
• On the Jeffreys prior for the multivariate Ewens distribution

Uses Software

• QRM

This page was built for publication: Bayesian estimation of a bivariate copula using the Jeffreys prior

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q418233)