贝叶斯推断下的条件风险价值研究  

Conditional Value at Risk Under Bayesian Inference

作　　者：王艳彩[1] 高岳林[1] 

机构地区：[1]北方民族大学信息与系统科学研究所,宁夏银川750021

出　　处：《河南科技大学学报（自然科学版）》2012年第6期91-94,10,共4页Journal of Henan University of Science And Technology:Natural Science

基　　金：国家社会科学基金项目(07XJY038)

摘　　要：基于假设的金融资产收益分布,建立了条件风险价值CVaR计算模型,并对分布的参数采用贝叶斯统计推断进行估计,然后利用蒙特卡洛模拟计算CVaR。选取沪深300中的股票数据进行实证分析,并与经典统计方法作比较研究,研究结果表明:应用贝叶斯推断估计的分布参数更精确,拟合的分布更符合数据的真实波动,提高了CVaR度量的准确性。Assuming financial assets income distribution was known,this paper proposed a new approach to conditional value at risk(CVaR) in framework of Bayesian inference.The parameters of the distribution were estimated via Bayesian inference,and CVaR was calculated by using Monte Carlo simulation.This new approach was tested on a set of stock data.The results indicate that the parameters estimated by Bayesian inference are more precise,which improves the measurement accuracy of CVaR consequently.

关 键 词：条件风险价值 贝叶斯推断 蒙特卡洛模拟 

分 类 号：F830[经济管理—金融学]