SAD and stock returns revisited: Nonlinear analysis based on MF-DCCA and Granger test

Abstract

Using both multifractal detrended fluctuation analysis (MF-DFA) and multifractal detrended cross-correlation analysis (MF-DCCA) methods, this paper examines the effects of seasonal affective disorder (SAD) on stock returns in the Chinese stock market. We first separate stocks into small- and large-cap stocks according to market value, and find that SAD and these two stock-return series are multifractal, based on MF-DFA. Moreover, we employ MF-DCCA and show that the cross-correlation between SAD and stock returns is anti-persistent, and the cross-correlation between SAD and return of small-cap stock is stronger than that between SAD and return of large-cap stock. Furthermore, we find that long-range correlation is the main source of multifractality in the cross-correlations between SAD and small-cap stock, while both long-range correlation and fat tails contribute to the multifractality between SAD and large-cap stock. Finally, utilizing nonlinear Granger Causality test, we find that SAD is the Granger cause of stock returns, not vice versa.