基于因子分析法的金融高频已实现波动的预测

金融高频数据的已实现波动(RV)在风险管理中扮演着非常重要的角色,已有大量文献对如何预测资产的已实现波动进行了研究.采用因子分析法来预测RV,探讨了不可观测的金融序列的公共因子在预测已实现波动时所起的作用,并考虑了资产价格中跳跃的影响,建立了基于因子分析法的波动预测模型(F-RV-J).从损失函数、MCS检验和在险价值VaR的预测能力三个方面,将F-RV-J模型与其它常用的预测模型进行了比较,发现F-RV-J模型明显要优于其它波动预测模型.

A Factor Approach to High Frequency Realized Volatility Forecasting

Realized volatility (RV) plays a very important role in risk management,there is a growing literature on the RV forecasting of asset returns using high frequency data.In this paper,we use the factor analysis method to predict RV,explore the role of common factor which reflect the unobservable financial characteristics,consider the impact of asset price jump,and establish the new F-RV-J volatility forecasting model.This paper compares the F-RV-J model with other forecasting model from loss function,MCS test and value at risk (VaR) aspects,and find that F-RV-J model is the optimal volatility forecasting model.