流动性视角下股指期货与股市已实现协方差预测:基于窗口平均支持向量回归方法  

The Prediction of Realized Covariance Between Stock Index and Futures Market From the Perspective of Liquidity:Based on Windows Average SVR Method

作　　者：曹杨丽 乔高秀[1] CAO Yang-li;QIAO Gao-xiu(School of Mathematics,Southwest Jiaotong University,Chengdu 611756,China)

机构地区：[1]西南交通大学数学学院,四川成都611756

出　　处：《数学的实践与认识》2021年第11期33-46,共14页Mathematics in Practice and Theory

基　　金：国家自然科学基金(71701171,72001180);教育部人文社会科学研究项目(17YJC790119)。

摘　　要：研究了流动性对我国股指期货与股票市场已实现协方差的影响,从一个新的角度揭示流动性对两个市场相关性的影响机制.基于沪深300指数和股指期货市场的高频数据计算已实现协方差,并提出基于支持向量回归(SVR)的窗口平均预测(AveW)方法研究已实现协方差的预测.实证研究发现,股市流动性对已实现协方差有显著的负向影响,基于SVR的窗口平均方法的预测效果显著优于传统的OLS和SVR估计方法,综合跳跃、共跳和流动性有助于获得更高的预测精度.This paper studies the influence of liquidity on the realized covariance of China’s stock index and futures markets,and reveals the mechanism of liquidity on the correlation between the two markets from a new perspective.Based on the high-frequency data of the CSI 300 index and index futures,the covariance is calculated,and the method of window average prediction(AveW) based on support vector regression(SVR) is proposed to study the covariance prediction.Empirical studies find that stock market liquidity has a significant negative impact on the realized covariance.The forecasting effect of the window average method based on SVR is significantly better than the traditional OLS and SVR estimation methods.Combining jumps,co-jump and liquidity help to obtain higher prediction accuracy.

关 键 词：流动性 已实现协方差 沪深300指数和股指期货 支持向量回归 窗口平均预测 

分 类 号：F832.5[经济管理—金融学]