Volatility persistence and trading volume in an emerging futures market

Purpose - The purpose of this paper is to estimate time‐varying conditional volatility, and examine the extent to which trading volume, as a proxy for information arrival, explain the persistence of futures market volatility using National Stock Exchange S&P CRISIL NSE Index Nifty index futures. Design/methodology/approach - To estimate the volatility and capture the stylized facts of fat‐tail distribution, volatility clustering, leverage effect, and mean‐reversion in futures returns, appropriate ARMA‐generalized autoregressive conditional heteroscedastic (GARCH) and ARMA‐EGARCH models with generalized error distribution have been used. The ARMA‐EGARCH model is augmented by including contemporaneous and lagged trading volume to determine their contribution to time‐varying conditional volatility. Findings - The paper finds evidence of leverage effect, which indicates that negative shocks increase the futures market volatility more than positive shocks of the same magnitude. In addition, the results indicate that inclusion of both contemporaneous and lagged trading volume in the GARCH model reduces the persistence in volatility, but contemporaneous volume provides a greater reduction than lagged volume. Nevertheless, the GARCH effect does not completely vanish. Practical implications - Research findings have important implications for the traders, regulatory bodies, and practitioners. A positive volume‐price volatility relationship implies that a new futures contract will be successful only to the extent that there is enough price uncertainty associated with the underlying asset. Higher trading volume causes higher volatility; so, it suggests the need for greater regulatory restrictions. Originality/value - Equity derivatives are relatively new phenomena in Indian capital market. This paper extends and updates the existing empirical research on the relationship between futures price volatility and volume in the emerging Indian capital market using improved methodology and recent data set.