高阶矩风险平价模型能否改善投资绩效?——来自中国市场的验证  

Can Risk Parity Model with Higher Moments Improve Investment Performance?——Evidence from the Chinese Stock Market

作　　者：于孝建[1,2] 陈曦[1] Yu Xiaojian;Chen Xi(School of Economics and Commerce,South China University of Technology,Guangdong Guangzhou 510006;Research Center of Financial Engineering,South China University of Technology,Guangdong Guangzhou 510006)

机构地区：[1]华南理工大学经济与贸易学院,广东广州510006 [2]华南理工大学金融工程研究中心,广东广州510006

出　　处：《金融发展研究》2018年第12期10-15,共6页Journal Of Financial Development Research

摘　　要：风险平价模型以资产波动衡量风险,忽略了资产收益分布的尾部特征。本文在风险平价模型中引入高阶矩风险,得到了九种不同的高阶矩风险平价模型,并选取平均相关性高的国内行业指数样本和平均相关性低的大类资产样本,对不同模型进行分析。研究发现:当标的资产间平均相关性较高时,包含偏度的高阶风险平价模型表现更优,投资组合的风险更小,收益更高;当标的资产间平均相关性较低时,风险平价模型表现更优。Risk parity approach measures risk by asset volatility,ignoring the tail behavior of the asset returns. This paper has compared nine risk parities with different higher moments' settings. Both industry index dataset with a higher average correlation and large class asset dataset with lower average correlation are studied with these models. The results show that when the average correlation between the underlying assets is high,higher-moment risk parity model with skewness performances better,and the portfolio has less risk and higher return;when the average correlation between the underlying assets is low,risk parity model performances better.

关 键 词：风险平价模型 高阶矩 相关性 

分 类 号：F832.5[经济管理—金融学]