Bayesian estimation of a bivariate copula using the Jeffreys prior

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.Comment: Published in at http://dx.doi.org/10.3150/10-BEJ345 the Bernoulli (http://isi.cbs.nl/bernoulli/) by the International Statistical Institute/Bernoulli Society (http://isi.cbs.nl/BS/bshome.htm

Grid-Uniform Copulas and Rectangle Exchanges: Bayesian Model and Inference for a Rich Class of Copula Functions

“…We compare our proposal with the flat prior, which is one of those proposed by Guillotte and Perron (2012). There are similarities between our model and their flat proposal.…”

“…Flexible model-based approaches for copula functions can be found in Guillotte and Perron (2012), Wu et al (2013a), Wu et al (2013b), and Ning and Shephard (2018). Guillotte and Perron (2012) proposed an interesting semi-parametric Bayesian approach for bivariate copulas based on a finite-dimensional approximation.…”

“…Flexible model-based approaches for copula functions can be found in Guillotte and Perron (2012), Wu et al (2013a), Wu et al (2013b), and Ning and Shephard (2018). Guillotte and Perron (2012) proposed an interesting semi-parametric Bayesian approach for bivariate copulas based on a finite-dimensional approximation. This approximation is structured from a partition of the unit interval based of intervals of the same length [(i − 1)/m, i/m], where m is the number of intervals and i ∈ {1, .…”

Bayesian estimation of generalized partition of unity copulas

“…In this paper we tackle the question how to t GPUCs resulting from a discrete partition of unity. [8] propose an MCMC based estimation approach suited for all copulas that can be parameterized by a doubly stochastic matrix. To t the copula, they indirectly draw the doubly stochastic parameter matrix from its Hilbert space representation in a random walk Metropolis-Hastings sampler.…”

Dependence Modelling with Copulas. By H.Joe. Boca Raton, Florida CRC Press. 2015. 480 pages. £ 57.99 (hardback). ISBN 978‐1‐4665‐8322‐1.

“…Arguably, this is a very narrow view of the relevance of Bayesian inference. Recent nonparametric Bayesian approaches to copula estimation as for instance by Guillotte & Perron () and Burda & Prokhorov () are not mentioned at all.…”