The main objective of this chapter is to estimate volatility patterns in the case of S&P Bombay Stock Exchange (BSE) BANKEX index in India. In recent past, the Indian banking sector was one of the fastest-growing industries and all major banks have been included in S&P BANKEX index as index benchmark constituent companies. The financial econometric framework is based on asymmetric GARCH (1, 1) model which is performed in order to capture asymmetric volatility clustering and leptokurtosis. Data time lag is considered from the first transaction day of January 2002 to last transaction day of June 2014. The empirical results revealed the existence of volatility shocks in the selected time series and also volatility clustering. The volatility impact has generated highly positive clockwise and impacted actual stocks. Moreover, the empirical findings reveal that the BANKEX index grown over 17 times in 12 years and volatility returns have been found present in listed stocks.
TopLiterature Review
Bollerslev (1986) has generalized ARCH model by including lagged valued of the conditional variance. The GARCH model allows a wider range of behavior and patterns especially in the case of more persistent volatility. The most general form of the model is GARCH (1,1) where GARCH stands for Generalized Autoregressive Conditional Heteroscedasticity. Moreover, a GARCH model or basically or Generalized ARCH model represents an extension of the ARCH model which otherwise is very similar to an ARMA model.
According to Brooks (1996) such a generalization of ARCH model, namely GARCH model can be perceived as an infinite order ARCH model. Engle (1982) argued that traditional econometric models assume a constant one period forecast variance and therefore in order to generalize this implausible assumption, it was implemented a new class of stochastic processes called autoregressive conditional heteroscedastic (ARCH) processes. On the other hand, Brooks (1996) suggested that it is highly unlikely that a GARCH model of order greater than one in the autoregressive and moving average components would be required, since by definition, a GARCH (1,1) model implies an infinitely long memory with respect to past news, i.e. innovations.
Prasanna and Menon (2013) suggested that the global process of financial integration between Indian stock market and international stock markets caused the absorption domestic and global news into the asset prices and stock indices. Trivedi (2013) argued that generally the banking sector in India was distinguished by accuracy and high confidence even in dramatic periods caused by extreme events such as the global financial crisis.
Nateson et al. (2013) investigated spillover effect of volatility in Indian BSE Sensex on BSE sectoral indices based on GARCH (1,1) model. The authors emphasized the importance of financial integration due to volatility transmission patterns especially in terms of risk management strategies and portfolio diversification for all investment sectors in India. Moreover, Basabi, Roy and Niyogi (2009) examined conditional volatility patterns of the BSE BANKEX index based on symmetric and asymmetric GARCH models. The authors revealed the existence of leverage effect considering the response to positive and negative news (innovations).
TopMethodological Approach
According to Birau, Trivedi and Antonescu (2014) the financial data series consist of daily closing asset prices for the selected stock index during the period between January 2002 and June 2014 with the exception of legal holidays or other events when stock markets haven’t performed transactions. The continuously-compounded daily returns are calculated using the log difference of the closing prices of stock market selected index, i.e. S&P Bombay Stock Exchange (BSE) BANKEX index as follows:
where p is the daily closing price.