Are MENA and Pacific Basin stock equity markets predictable?

Abstract

This research uses variance ratio analysis to test whether Middle Eastern, North African (MENA) and Pacific Basin emerging equity markets follow a martingale behavior during the period1980-2004. The conventional Lo and MacKinlay variance ratio test, the multiple variance ratio test of Chow and Denning, rank- and sign-based test of Wright, and wild bootstrap of Kim are used for the monthly return series. The problem of thin trading was addressed using Miller, Muthuswamy, and Whaley’s adjusting procedure. Results have shown traces of a martingale behavior at high holding horizons. However, overall conclusions indicate that the null martingale hypothesis is strongly rejected for the whole sample and considered sub-periods at a 5% significance level. The pattern of the variance ratio estimates signify that the selected stock markets exhibit persistent mean-reverting and predictable behavior in their monthly adjusted returns series. The results expose the ineffectiveness of economic liberalization and privatization measures implemented in the early 1990s to improve their market efficiency. The Asian crisis did not affect the outcomes of the variance ratio analysis. Moreover, it sounds as if the perceptible development in terms of size and liquidity was not sufficient to exhibit a martingale behavior in these markets.