商品期货投资组合与市场收益的尾部相依研究--基于混频分位点相协模型  

The Tail Dependence Between Commodity Futures Portfolios:Based on qpr-MIDAS Model

作　　者：郭冉冉 叶五一[2] 刘小泉 缪柏其[2] Guo Ranran;Ye Wuyi;Liu Xiaoquan;Miao Baiqi(School of Economics,Hefei University of Technology,Anhui,Hefei 230601,China;School of Management,University of Science and Technology of China,Anhui,Hefei 230026,China;Nottingham University Business School China,Zhejiang,Ningbo 315100,China)

机构地区：[1]合肥工业大学经济学院,安徽合肥230601 [2]中国科学技术大学管理学院,安徽合肥230026 [3]诺丁汉大学商学院(中国),浙江宁波315100

出　　处：《中国管理科学》2024年第10期11-19,共9页Chinese Journal of Management Science

基　　金：国家自然科学基金青年项目(72201258);国家自然科学基金面上项目(72371230);安徽省杰出青年基金项目(2208085J41)。

摘　　要：资产的多样化投资是投资管理的核心话题,风险因子投资组合是投资者进行资产配置选择的重要参考,具有重要的研究意义。结合混频数据模型的思想,本文拓展并构建了qpr-MIDAS(quantile-specific probability ratio-Mix Data Sampling)分位点相协模型。文章基于中国商品期货市场,利用日高频商品指数以及股票指数,分别刻画了周频换手的风险因子与市场因子间的极值风险相依度量。从模型角度来看:结合MIDAS思想的qpr模型能够有效地刻画尾部相依度量,模型拟合效果更优;在经济应用中,该模型能够更有效地实现投资组合的多样化配置。同时,实证结果表明:(1)风险因子组合与市场因子收益间风险溢出关系相对独立;(2)因子组合间风险溢出的效果与商品指数以及股指收益存在明显的负向联动关系;(3)风险的传导效应会随市场准入政策放宽而减弱。本文的研究可为投资者在考虑极端风险情况下的商品期货资产组合配置以及投资组合设计提供理论借鉴。The diversified allocation of asset portfolios interested many scholars and investors.Additionally,diversifying existing risk factor portfolios is a common strategy for portfolio allocation.However,diversified allocation of factor strategies may not be as advantageous when there is a high correlation between factors.Meanwhile,the Chinese commodity futures market is highly active,and its unique market environment causes different manifestations of risk factors.It is interesting to investigate the correlation between different factors in the Chinese commodity market.It focuses on whether commodity futures factors are applicable in the Chinese futures market and whether the allocation of risk factors can help investors optimize assets.The qpr quantile association model is employed to effectively capture tail dependence.Building upon the work of Li Ruosha et al.,the application of covariates is further expanded upon by integrating the mixed data model MIDAS(Mixed-Data Sampling Model).Given the relatively short historical data available in China's financial market,the utilization of the MIDAS method becomes crucial.By amalgamating high-frequency daily data,this approach not only enriches the available data information but also enhances the model's fitting capabilities.In this paper,daily high-frequency commodity index and stock index data are used to describe the weekly tail dependence between risk factor returns and market return.Regarding the model:1)Comparing the basic qpr model,the qpr-MIDAS model can better fit the tail dependence results;2)the model can realize the diversification return with the allocation of investment portfolios in economic applications.The empirical results show that:1)there is no obvious spillover relationship between risk factor combinations and market returns except during crises;2)among all the factors,the momentum factor performs best,with high portfolio return and with low correlation with the market factor;3)the risk spillover effect negative link with the returns of commodity indexes

关 键 词：商品期货投资组合 尾部相依性 MIDAS 分位点相协回归模型 

分 类 号：F830[经济管理—金融学]