English

Adaptive estimation of the copula correlation matrix for semiparametric elliptical copulas

Machine Learning 2016-02-16 v3

Abstract

We study the adaptive estimation of copula correlation matrix Σ\Sigma for the semi-parametric elliptical copula model. In this context, the correlations are connected to Kendall's tau through a sine function transformation. Hence, a natural estimate for Σ\Sigma is the plug-in estimator Σ^\hat{\Sigma} with Kendall's tau statistic. We first obtain a sharp bound on the operator norm of Σ^Σ\hat{\Sigma}-\Sigma. Then we study a factor model of Σ\Sigma, for which we propose a refined estimator Σ~\widetilde{\Sigma} by fitting a low-rank matrix plus a diagonal matrix to Σ^\hat{\Sigma} using least squares with a nuclear norm penalty on the low-rank matrix. The bound on the operator norm of Σ^Σ\hat{\Sigma}-\Sigma serves to scale the penalty term, and we obtain finite sample oracle inequalities for Σ~\widetilde{\Sigma}. We also consider an elementary factor copula model of Σ\Sigma, for which we propose closed-form estimators. All of our estimation procedures are entirely data-driven.

Keywords

Cite

@article{arxiv.1305.6526,
  title  = {Adaptive estimation of the copula correlation matrix for semiparametric elliptical copulas},
  author = {Marten Wegkamp and Yue Zhao},
  journal= {arXiv preprint arXiv:1305.6526},
  year   = {2016}
}

Comments

Published at http://dx.doi.org/10.3150/14-BEJ690 in the Bernoulli (http://isi.cbs.nl/bernoulli/) by the International Statistical Institute/Bernoulli Society (http://isi.cbs.nl/BS/bshome.htm)