主 题:Empirical Likelihood Based Goodness-of-Fit Test for Copulas
内容简介:Copula method is now widely used in finance, particularly when modelling dependence structure of several variables is needed. However, in practice how to evaluate the fitness of a specific parametric copula to the data is to be carefully considered. In this paper we develop a goodness-of-fit test based on empirical likelihood (EL). Moment conditions that contain information about data and parameters play a vital role in calculating the EL, and here we derive the moment condition for the EL test based on kernel smoothed empirical copula. The novelty of our approach is that we incorporate the dependence structure of data in the moment conditions, which is the main purpose of fitting data a copula. We show that the test statistic under null hypothesis follows a chi-square distribution and simulation results to be provided.
报告人:王星 副教授
时 间:2017-06-29 10:00
地 点:竞慧东楼302室
举办单位:理学院