基于周内多重分形波动率的基金投资风格漂移风险测度研究

基金发生投资风格漂移是把双刃剑，在获得短期超额收益时，也隐藏着巨大的风格漂移风险。本文首先以我国79只开放式股票型基金为样本，在量化投资风格漂移的基础上，分析发现其收益序列存在多重分形特征，据此构建周内多重分形波动率测度来刻画投资风格漂移收益的复杂波动特征，并与传统的GARCH族波动率计量模型的测度能力进行比较分析，实证结果发现本文构建的周内多重分形波动率测度更加精确，能更好刻画序列的复杂波动特征；然后，进一步构建MFVW VaR模型对基金投资风格漂移风险进行量化测度，发现该模型比传统的参数与非参数VaR模型能更好地对风格漂移风险进行有效测度，基金普遍存在较大的风格漂移风险；最后，对我国开放式股票型基金的产品创新策略与投资风格漂移监管策略进行了一些有益探讨。

Measuring the Investment Style Drift Risk of Open-end Equity Funds Based on Multifractal Volatility in Week

Fund investment style drift is a double-edged sword: high short-term excess returns and enormous drift risk. This paper constructs the MFVW-VaR model to measure 79 open-end equity funds’ drift returns in Chinese fund market and characterize the multiple features of style drift return. Compared with RiskMetrics model and GARCH family models, MFVW model can be more accurate and stable; moreover, MFVW-VaR can measure the investment style drift risk more effectively than GARCH-VaR. Therefore, this new method can accurately calculate the style drift risk and provide precious suggestions for supervision on open-end equity fund market in China.