Volatility Indices and State-preference Pricing

This paper is the first to use a state-preference pricing approach with BlackScholes analytic second derivatives to develop a forward-looking volatility index (FIX), as a forecast of the next 30-day market risk-neutral volatility. Using S&P 500 index (SPX) option prices from 1996 to 2010, we find that FIX is 99% correlated with the current CBOE volatility index (VIX) and it is a better estimator for the shortterm realized volatility of SPX returns than VIX. Our result is robust to different measures of realized market volatilities. We argue that VIX is not a model-free measure and due to the potential noise issues from the thinly traded deep out-ofthe-money (OTM) options, VIX may over-estimate the future volatility. This is supported by our results. Moreover, unlike VIX which is affected by the availability of strike prices and may be manipulated by trading deep OTM put options, we show that FIX is more difficult for manipulation and less data dependent. We also show that FIX provides a better volatility forecast than other alternative measures, such as the squared root of GARCH (1, 1) variance. Our results reinforce previous findings in the literature that at-the-money Black-Scholes implied volatility is an efficient and more superior forecast of subsequent realized volatility. ∗3/52 York Street, Indooroopilly, QLD, Australia, 4068. E-mail: [email protected]. I would like to thank Professor Tom Smith for his ongoing encouragement, support and inspiration throughout the preparation of this paper. All errors and omission are mine.