The importance of return volatility forecasts in policy formation and investment decision-making in emerging countries is growing considerably. However, from an operational perspective, there is no consensus in the literature on which econometric model has the best forecasting performance. To shed new light on this issue, this article compares forecasting models for a selected group of emerging Asian economies: India, Malaysia, Pakistan, Sri Lanka, Singapore and Thailand. Model’s performance is tested using both in-sample and out-of-sample forecasting methods. It is found that a relatively simple asymmetric EGARCH model clearly outperforms other models. JEL Classification: G12, G17
This article examines as to whether past stock prices in the Colombo Securities Exchange (CSE) exhibit predictability of future prices by using various statistical tests for the period from 1985 to 2009 on daily data. The findings of various statistical tests generated mixed results. The results of tests on serial correlation, runs test, variance ratio test and tests on the day-of-the-week effect rejected the weak-form of Efficient Market Hypothesis, whereas Augmented Dickey-Fuller (ADF) test and Phillips Perron (PP) unit root tests as well as the month-of-theyear test failed to reject the presence of random walk. Findings of the ADF and PP test statistics implied that daily changes in stock prices cannot be predicted, so that no investor can earn abnormal profits by exploiting past stock price patterns. This indicates the presence of weak-form efficiency in CSE. However, the results of serial correlation tests, runs test and variance ratio test imply that prices of ASPI do not reflect past prices, indicating that future movements of the stock prices can be predicted by using past price movements. The idea that past price information can only partially predict the future movements of prices/ returns indicates that other information is also vital in changing stock prices/ returns. The article concludes that the weak-form inefficiency is the normal situation in the CSE.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.