中文摘要：

首先应用结合变量选择的logistic模型估计了上市公司的违约概率，接着采用核密度估计方法和Archimedeancopula函数分别拟合了上市公司的违约概率变化和市值变化的边缘分布和联合分布．并在大样本下通过条件VaR检验了结构化信用分析模型在中国市场的适用性．实证结果显示上市公司信用风险和公司市值变化之间存在负相关关系，公司财务的恶化或好转会对市值产生影响，且违约概率上升风险对市值降低较为敏感．此外，市场违约概率分布的变化是影响模型预测精度的主要因素，而违约概率和市值变化间相依结构的影响则相对较小．

英文摘要：

Based on a variable selection method, the default probability of listed companies was calculated with a logistic model. Then kernel estimation and copula method were applied to fit the marginal and joint distribution between credit risk and changes in market value of listed companies. With conditional VaR, the applicability of the structure credit risk model was tested with large samples. Empirical tests show that there is a negative relationship between credit risk and market value which is affected by financial ratios and the default probability is sensitive to the decrease in market value. Besides, the main factor in forecast accuracy is the distribution of default probability, while the impact of the dependence structure between default probability and market value is not significant.