高频数据下波动择时在动态资产配置中的应用研究

The Application Research of Volatility Timing in Dynamic Asset Allocation Based on High Frequency Data

文章研究了市场微观结构噪音条件下波动择时在动态资产配置中的应用,使用高频数据在均值-方差投资组合框架下分析了波动性估计精度提高对资产配置的影响。研究结果表明,与使用固定采样频率(5分钟)相比,使用根据最优采样频率计算的已实现方差/协方差进行动态资产配置可以获得更高的经济收益,表明在高频数据条件下考虑市场微观结构噪音进行波动性估计具有重要的意义。

This article studies the application of volatility timing in dynamic asset allocation based on market microstructure noise.The effect of improving volatility estimates on asset allocation is analyzed under a framework of mean-variance portfolio.The results shows that a risk averse investor who is given the option of choosing variance/covariance estimates derived from optimal sampling methods versus estimates obtained from 5-minute intervals will achieve more economic gains,indicating that based on high frequency data volatility estimation in the presence of microstructure noise is valuable.