基于门限分位点回归的条件VaR风险度量  

Computing Conditional VaR Based on Threshold Quantile Regression

作　　者：余力[1] 张勇[1] 李国勇[1] 

机构地区：[1]西安交通大学经济与金融学院,西安710061

出　　处：《经济问题》2010年第4期104-108,共5页On Economic Problems

基　　金：西安交通大学"985"工程二期建设项目(07200701)的阶段性成果

摘　　要：门限分位点回归模型是线性分位点回归模型的改进,是一种更加客观实际的非线性估计方法。利用该模型实证分析了单只股票(民生银行)的条件VaR。选择一种流动性风险指标作为条件,经过分析发现,由门限分位点回归模型得到结果能够更好地描述市场,也能更好地预测市场风险。In this paper,the threshold quantile regression model is presented,which is revised model of the linear quantile regression,which can not suit practical demand.an empirical analysis on Minsheng Bank stock based on this model is given.and a liquidity risk measure is picked as the condition variable.By the analysis,the result obtained based on threshold quantile regression model can describe the practical market better,and forecast the market risk better.

关 键 词：门限分位点回归模型 分位点回归模型 条件CVaR 

分 类 号：F83[经济管理—金融学]