中文摘要：

本文对不同族类、不同种类Copula函数进行比较分析，基于深圳发展银行、上海浦东发展银行、中国民生银行信用收益率和汇率收益率的经验数据进行实证研究与分析，用经验累积分布函数拟合信用收益率和汇率收益率的边缘分布，ArchimedeanCopula函数描述三家银行的信用收益率和汇率收益率的相关关系。在各个置信度下，运用拟合度最优的Copula函数求出各自的信用-汇率风险的VaR（Value—at—risk）值和CVaR（Condifonal Value-at-risk）值，并比较三家银行整合风险的大小，为金融机构决策者的监管提供参考。

英文摘要：

This article compared and analyzed copulas from different families and types. The test and analysis were on empirical data of Shenzhen Development Bank,Shanghai Pudong Development Bank and China Minsheng Banking CORP.LTD, we use a method of empirical cumulative distribution function to fit its unknown marginal distribution, Archimedean Copula to characterize the relationship between this three banks＇ credit yield and return rate of exchange rate. And then we use Gumbel Copula to calculate the VaR and CVaR of Credit-exchange rate Integrated Risk under each Confidence level. After that, we conduct comparative analyses among these three banks and try to provide reference for financial institute decision -makers in supervising.