Non Linearity and Heteroskedasticity Effect on Stock Returns Volatility

This article investigates the asymmetric nature of US stock market return and effect of heteroskedasticity on stock return volatility. Further, this study also analyzes the relationship between stock returns and conditional volatility, and standard residuals. The monthly opening and closing prices of S & P 500 are used for the period from January 1950 to December 2007. The study applies Ljung-Box statistics to examine the autocorrelation in stock returns and GARCH (1, 1) and TAR-GARCH (1, 1) to test the heteroscedasticy, and asymmetric nature of stock returns respectively. The results of the study suggest the presence of non linearity, heteroskedasticity effect and asymmetric nature of stock returns. It also finds no correlation between stock returns and conditional volatility, however, the relationship between stock returns and standardized residuals is found positively significant. These findings bring out the essential elements of the modern investment theory that investors adjust their investment decisions with respect to expected volatility, however, they tend to earn extra risk premium for unexpected volatility. This study provides a robustness test of the conditional volatility and asymmetric impact of good and bad news.