Does One Size Fit All? Evidence from forecasting Stock Market Returns in India

Abstract

The paper studies the volatility patterns in Indian markets using National Stock Exchange based CNX-500 Index between 1999 and 2013. The index represented 96.76 per cent of the total market capitalization of the National Stock Exchange (NSE) as in December, 2013. The objective of this paper is two-fold. At a broader level, it studies stock returns in India with a focus on the characteristics of returns and the volatility patterns displayed by them. At a more specific level, based on the characteristics of stock returns that emerge, an attempt has been made to locate an appropriate forecasting tool for the Indian stock market. In the analysis of market returns, there are evidences of the presence of ‘volatility clustering’, ‘leverage effect’ and ‘stationarity’ in the return series with a possibility of unit root in the variance of the return series. After modeling the volatility patterns exhibited, the findings suggest that no one model emerges as the best forecasting model and managers must choose models appropriate to their requirements.