Stock Returns Memories: a “Stardust” Memory?

This article aims at investigating econometrically the market efficiency concept through an analysis of the dependence structure of stock market index returns. To that purpose, we use a large range of methods in this paper. Six different estimation procedures are applied to obtain the Hurst exponent, starting with the “R/S” approach, continuing with ARFIMA models and ending with wavelet models. We investigate the possible presence of long or short-memory in twelve market indexes between three periods, namely (1960-2013), (1980-2013) and (1990-2013). Our conclusions depend on the degree of financial maturity: most emerging markets display the presence of memory, whereas mature markets show an absence of or very short-memory dynamics.