题 目：Selection of Copulas with Applications in Finance?

汇报人：Zongwu Cai教授

a Department of Mathematics & Statistics,

University of North Carolina at Charlotte,

Charlotte, NC 28223, USA, E-mails:zcai@uncc.edu

The Wang Yanan Institute for Studies in Economics,

Xiamen University, China

时 间：2008年5月26日(周一)上午10：00

地 点：J9集团国际站219

Abstract

A fundamental issue of applying copula method in applications is how to choose an appropriate copula function. In this article we address this issue by proposing a new copula selection approach via penalized likelihood. The proposed method selects the appropriate copula functions and estimates copula coefficients simultaneously. The asymptotic properties, including the rate of convergence and asymptotic normality and abnormality, are established for the proposed penalized likelihood estimator. Particularly, when the true coefficient parameters may be on the boundary of the parameter space and the dependence parameters are in an unidentifiable subset of the parameter space, it shows that the limiting distribution for boundary parameters is abnormal and the penalized likelihood estimator for unidentified parameters converges to an arbitrary value. Moreover, the EM algorithm is proposed for optimizing penalized likelihood function. Finally, Monte Carlo simulation studies are carried out to illustrate the finite sample performance of the proposed method and the proposed method is used to investigate the correlation structure and comovement of financial stock markets.