The impulsiveness in investment’s price is volatility and its meticulous estimation and forecasting is valuable to investors in the risk management of their portfolio. Earlier volatility of an asset was assumed to be constant. However, the pioneering studies of Mandelbrot, Engle and Bollerslev on the property of stock market returns did not support this assumption. The family of autoregressive conditional heteroskedasticity models were developed to capture time-varying characteristics of volatility. The present treatise attempts to study the presence of autoregressive conditional heteroskedasticity in four Indian banking sector indices viz. BSE Bankex, BSE PSU, CNX bank and CNX PSU. The daily banking sector indices for the period of January 2004 to December 2013 were taken from the online database maintained by the Bombay Stock Exchange and the National Stock Exchange. The data of four indices was studied for stationarity, serial correlation in the returns and serial correlation in the squares of returns with the help of Augmented Dickey–Fuller test, Box-Jenkins methodology and autoregressive conditional heteroscedasticity models respectively. The results of ACF, PACF and Ljung–Box Q test indicates that there is a tendency of the periods of high and low volatility to cluster in the Indian banking sector. All the four banking sector indices display the presence of ARCH effect indicating the presence of volatility clustering. Engle's ARCH test (i.e Lagrange multiplier test) and Breush-Godfrey-Pagan test and ARCH model confirmed the high persistence and predictability of volatility in the Indian banking sector.
Testing For Differences Between Two Distributions In The Presence Of Serial Correlation Using the
Kolmogorov–Smirnov
and Kuiper's Tests
