Predictability of multifractal analysis of Hang Seng stock index in Hong Kong

In this paper, the daily Hang Seng index in Hong Kong stock market is studied by multifractal analysis. The main parameter of multifractal spectra used is Δf, which can be used to characterize the ratio of number of highest index moments to that of lowest ones. The dependence of todayʹs gain probability (G%) and the dayʹs index increase probability (n%) with Δf of the previous 3 days are studied. It is found that G% and n% can reach 70–80% at the large positive Δf region and can be very close to 20% at the big negative Δf region. The predictability decreases with the increasing number of the previous days.