Day-of-the-Week Effect: An Enduring Anomaly in Bursa Malaysia

(ProQuest: ... denotes formulae omitted.)IntroductionSince Fama coined the term 'efficient market' in 1970, there has been numerous attempts to sustain the Efficient Market Hypothesis (EMH) which emphasizes that the asset prices fully reflect 'available' information. However, these attempts find no evidence in favor of EMH (Islam et al., 2007). The EMH holds that as a result of competition, equilibrium prices in financial markets incorporate all relevant information. Hence, simple strategies cannot beat stock markets because the prices follow a random walk. If the stock prices follow a random walk, the market is said to be efficient and the prices are not predictable based on the past prices, as the price would be dependent upon the new information. The fact that the return volatility is not independent of the time (i.e., time variant; the presence of calendar anomalies) evidences against random walk theory. If the information flows efficiently, investors share similar information creating no systematic way of exploiting trading opportunities, but an investor would outperform using unique strategies if anomalies exist. Therefore, empirical studies on seasonality are at least motivated by (1) finding evidence for EMH; and (2) information for judgments of managers and investors.Financial theory issues such as volatility, predictability, speculation and anomalies are also related to the efficiency issue (Islam et al., 2007) and are all interdependent. Therefore, EMH tests have taken the form of testing the issues of predictability, anomaly, seasonality, volatility and the existence of bubbles. Seasonal, momentum and disposition effects in markets are some of the contrary evidences that have been the outcome of such empirical efforts received by EMH. Some studies suggest that the limited support for EMH is mostly due to inappropriate test methods, excess volatility, mean reversions, nonlinearity, anomalies and bubbles in the stock markets. Chien et al. (2002) state that the limitations of the existing method of efficiency tests contribute to the conflicting outcomes. The use of inappropriate statistical methods to test the seasonality in stock market makes the evidence on seasonality an 'invalid challenge' for EMH. In contrast, a number of studies, including that of French (1980), Harris (1986), and Nippani and Greenhut (2011), confirm the distribution of stock returns according to the day-of-the-week. Evidence on day-of-the-week effect also suggests that behavioral factors would be priced and makes invalid the 'Homo Economicus' assumption that the humans make perfectly rational decisions. Research on seasonal effects suggests the existence of crowd behavior in markets even though the markets are efficiently adjusted to new information. Hence, price adjustments do not show rational economic decisions and it is not the weaknesses of test methods but irrationality of the market.Day-of-the-week effect is the most widely documented among the seasonal effects and has been comprehensively investigated in different markets (Rahman, 2009). Basdas (2011) observes that studies generally adopt regression analysis with dummy variables, ANOVA, non-parametric tests or generalized autoregressive modeling for testing seasonality effect. However, Chien et al. (2002) suggest non-parametric tests to avoid upward biasness in test statistics derived from dummy variable models, particularly when the return volatility of a period observed is higher than the other periods. The present study uses both parametric and non-parametric statistics in examining the seasonality. If the day-of-the-week effect is true in an efficient market, the anticipations would make it difficult to create profit out of seasonality. Nippani and Greenhut (2011) find a reversal of weekend effect after a long period of weekend effect, suggesting that the market follows seasonality in making money. According to Muhammad and Rahman (2010), Malaysian Stock Exchange has shown day-of-the-week effect in the past. …