Estimating conditional jumps in volatility using realized-range measures

We propose an extension of HAR model for estimating the presence of jumps in returns volatility, using the realized-range measure as a volatility proxy. We show that, once that squared jumps in prices are disentangled from integrated variance, then there is a positive probability of jumps in volatility, conditional on the past information set. The analysis carried out with 16 NYSE stocks highlights the contribution of jumps during periods of financial turmoil. The common component of the estimated expected jumps, across the stocks and obtained by principal components, turns out to be correlated with a set of lagged financial variables, where the most significant is the credit-default swap index on US banks. The estimation results of extended HAR-V-J model, modified to incorporate the information content of credit-default swap in the dynamics of the jump size and intensity, confirm the significative contribution of the credit-default swap to the dynamics of the jump size.