• C22;
  • G14
  • Fractional integration;
  • mean reversion;
  • random walk;
  • ARFIMA model;
  • Wavelet analysis;
  • Markov model


This paper expands and augments the results of the paper by Jefferis and Thupayagale) and tests the efficiency of the South African stock market with Wavelet and Markov Switching Regime analyses of selected shares and the a ALSI 40 data. The Wavelet analysis indicated that most of the individual share prices and the share index time series are mean reverting over the long run and follow a long memory process, offering evidence against weak-form efficient market hypothesis (EMH). The Markov model modelled the financial and prevalent economic conditions accurately and established the presence of patterns in the historic time series, providing additional support against the weak-form EMH.