Multifractal characteristics and return predictability in the Chinese stock markets

TL;DR

This paper uses multifractal analysis to reveal complex market dynamics in Chinese stock indices and demonstrates that multifractal spectral features can predict future returns, offering new insights into asset pricing.

Contribution

It introduces the application of multifractal detrended fluctuation analysis to Chinese stock markets and shows spectral width as a significant predictor of returns.

Findings

Multifractal spectra characterize market volatility.

Spectral width predicts future excess returns.

Significant relationship between spectral features and returns.

Abstract

By adopting Multifractal detrended fluctuation (MF-DFA) analysis methods, the multifractal nature is revealed in the high-frequency data of two typical indexes, the Shanghai Stock Exchange Composite 180 Index (SH180) and the Shenzhen Stock Exchange Composite Index (SZCI). The characteristics of the corresponding multifractal spectra are defined as a measurement of market volatility. It is found that there is a statistically significant relationship between the stock index returns and the spectral characteristics, which can be applied to forecast the future market return. The in-sample and out-of-sample tests on the return predictability of multifractal characteristics indicate the spectral width $\Delta {\alpha}$ is a significant and positive excess return predictor. Our results shed new lights on the application of multifractal nature in asset pricing.