Multifractal analysis on stock style asset returns based on sliding windows MF-DFA

作者：Xu, Lin^*; Song, Guang-Hui; Guo, Wen-Wei

来源：Systems Engineering-Theory & Practice, 2012, 32(9): 1891-1899.

• Generalized Hurst exponent
• Multi-fractal spectrum
• Multifractal characteristics
• Multifractal detrended fluctuation analysis
• Position reconstruction
• Sliding Window
• Sliding window techniques
• Stock styles

Style investment has become a main quantitative investment approach to construct investment portfolio for funds. This paper analyzed the fractal characteristics of stock style assets returns, and revised the multifractal detrended fluctuation analysis through introducing a sliding window technique, researched on the daily return volatility characteristics of six kinds of pure stock style assets indexs introduced by Standard & Poor company, the empirical results showed that: sliding window technique can effectively reduce the fake-flutrated errors from the unconuity on junction points; daily return series of style asset indexs have relevant multifractal characteristics, the characteristics mean that the original style asset series is persistent and position reconstruction series have anti-persistence, the multifractal characteristics of position reconstruction series were significantly weaker than the multifractal characteristics of original series, which indicates enduring relevance of style asset indexs return series was the main reasons of multifractal characteristics; value & growth syle assets have stronger multifractal characteristics than the size style assets, indicating that the multi-fractal characteristics of size style assets were significantly weaker than the multifractal characteristics of value & growth syle assets. There is an important theoretical value and practical significance for fund managers timely and accuratly grasp the market style trends to construct a moderate style drift strategy.

• 出版日期2012-9
• 单位华南理工大学