中国股市经流动性调整的极值风险测度研究  被引量:2

Study on Extreme Risk Measurement of Chinese Stock Market with Liquidity Adjusted

在线阅读下载全文

作  者:覃小兵 陈粘 林宇[1] 

机构地区:[1]成都理工大学商学院 [2]成都理工大学管理科学学院

出  处:《金融理论与实践》2015年第10期81-88,共8页Financial Theory and Practice

基  金:国家自然科学基金资助项目(71171025;71271227;71371157);国家社会科学基金资助项目(12BGL024);四川省软科学研究计划资助项目(2014ZR0093);四川省创新创业计划资助项目(201410616031);成都理工大学"金融与投资"优秀创新团队计划资助项目(KYTD201303)

摘  要:针对现有市场极值风险测度方法不能反映市场流动性风险的缺陷,采用价量结合的风险测度方法,引入由换手率(Turnover Rate,以下简称TR)求得的变现时间和极值理论(Extreme Value Theory,以下简称EVT)对Bangia等(1998)提出的经流动性调整Va R的La-Va R模型(BDSS模型)进行改进,进而运用改进后的La-Va R模型对上证综指(Shanghai Stock Exchange Composite Index,以下简称SSEC)经流动性调整的极值风险进行测度,并采用规范的返回测试检验方法(Back-testing)对模型的稳健性进行检验。实证结果表明:改进后的La-Va R模型比传统的BDSS模型更能准确测度中国股市经流动性调整的极值风险;在改进后的La-Va R模型中,基于极值理论的La-Va R模型比基于学生t分布的La-Va R模型不仅更能准确地测度风险,而且模型的溢出情况也更为随机,从而拥有更好的稳健性。Aiming at the issue that the existing market extreme risk measure methods can’t reflect the market liquidity risk, in this paper, we use liquidation time and extreme value theory (EVT) to modify the BDSS model, a kind of VaR model with liquidity adjusted which was put forward by Bangia et al in 1998.And then we use the improved La-VaR model to measure the extreme risk of shanghai stock exchange composite index (SSEC) with liquidity adjusted. Besides, we also use the normative back-testing method for the robustness testing of the La-VaR models. The empirical results show that the improved La-VaR models are better than the traditional BDSS model in the measurement of extreme risk in Chinese stock market with liquidity adjusted;and among the improved La-VaR models, the La-VaR model based on EVT not only has a better ability to measure risk accurately, but also can get a much more random situation for the failure prediction than the La-VaR model based on student t, so the La-VaR model based on EVT has a better performance.

关 键 词:股票市场 BDSS模型 极值理论 流动性调整 变现时间 风险测度 

分 类 号:F830.91[经济管理—金融学]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象