The forecasting performance of implied volatility index: evidence from India VIX

Abstract

In this paper, we investigate the forecasting performance of ex-post an ex-ante volatility forecasts against realized return volatility of various time horizon. The competing volatility forecasts are implied volatility, RiskMetrics and GJR-GARCH; the empirical results uncover that implied volatility dominates the other volatility forecast in the prediction of future realized return volatility. The in-sample forecast suggests that ex-ante volatility best explains the future market volatility. The non-overlapping sampling procedure gives the more robust estimate of volatility forecasts, the results reveals that implied volatility forecasts of all horizon appears positive unbiased forecaster of realized volatility. Moreover, the instrumental variable estimation in the presence of error-in-variable clears that implied volatility is free from measurement error; OLS estimates remains more consistent than the 2SLS estimates. The information content of implied volatility encourages the exchanges to construct the implied volatility indices and volatility products on underlying volatility index.