edf40wrjww2CF_PaperMaster:Desc
Predictability of stock price returns in select emerging markets
A question that financial economists have tried to answer for many years is whether stock returns are predictable. Predictability of stock price returns is associated with the Efficient Market Theory and the Random Walk Hypothesis (RWH). The Efficient Market Theory and the RWH postulate that the current stock prices fully reflect the available information and the successive stock prices are independent of each other. Current stock prices, therefore, are unpredictable. These two have been the subject of debate in financial literature. A considerable amount of research has been carried out in examining the efficiency of stock market price formation. A vast majority of these studies were unable to reject the efficient market theory as reported by Fama’s (1970) survey.
Subsequent empirical research (Fama & French,1988; Poterba & Summers,1988; Lo & Mackinlay,1988; Urruti,1995), however, provides evidence that stock prices do not follow a random walk. Several empirical tests are available to test the RWH. The traditional tests of the RWH include the Runs Test and the Serial Correlation Test. The Runs Test is non-parametric and the Serial Correlation Test assumes normality and homoscedasticity of return distribution. The Variance Ratio (VR) Test proposed by Lo and MacKinlay (1988) addresses the problem of non-normality and heteroscedasticity in financial time series. There is ample evidence that shows that financial time series possess time-varying volatilities, and, therefore, deviate from normality. Ayadi and Pyun (1994) acknowledge that the variance ratio test is more appealing than the other traditional tests of random ...