UC Santa Cruz
Essays On The Persistence Of The Forecast Bias Of Option Implied Volatility
- Author(s): Asensio, Ivan Oscar
- Advisor(s): Hutchison, Michael
- et al.
Chapter I contains a literature review on the forecast bias of implied volatility based on the two fundamental questions addressed in the literature. Does implied volatility evolve as suggested by its term structure? And, is ex post realized volatility consistent with ex ante implied volatility forecasts? Chapters II and III contribute to this literature by focusing on explanations for and interpretations of forecast bias persistence. Some literature review sections which deal with specific applications of bias persistence were embedded in Chapters II and III, instead of the literature review in Chapter I, so as not to disrupt the flow of this dissertation.
The first strand investigates the evolution of implied volatility itself across time. Application of the expectations hypothesis suggests that the shape and slope of today's implied volatility term structure should be consistent with expected future changes of short-dated implied volatility. Analysis of the information content in VIX (CBOE Volatility Index) futures in Chapter II reveals a persistent forecast bias. I find that while there is evidence to uphold expectations hypothesis during the 2008-9 credit crisis period and before, in general ex ante forecasts of the future level of the VIX, implied by the VIX term structure, overshoot actual ex post changes, especially over shorter tenors. Strategies designed to profit from the bias reported have not been successful in eliminating the arbitrage opportunity. The forecast bias has increased as additional capital has flowed in, a result which does not support the slow-moving capital explanation of arbitrage persistence. Instead, I present evidence to suggest that the VIX-VIX Futures Puzzle is propagated by significant inflows of capital from non-professional investors via the proliferation of ETF (exchange-traded fund) offerings.
The second fundamental question addressed in this literature investigates the extent to which implied volatilities accurately forecast the ex-post realized volatility of underlying asset prices. In other words, are option premiums justified by the subsequent payouts? Chapter III is devoted to this question, based on an expansive dataset of implied volatility for over thirty currency pairs across developed and emerging economies. I report the forecast biasness and propose that the magnitude of such may be used as a proxy for the degree of financial integration achieved for a particular country. I motivate this concept by furnishing a simple theoretical framework based on Dornbusch, Fischer, Samuelson (1977), based on the foundation that currencies with lower (greater) cross-border trading frictions would be expected to have options markets that exhibit lower (higher) levels of forecast bias. The transmission mechanism from cross-border currency trading frictions to FX implied volatility forecast bias is the hedging activity of currency option market-makers, based on the Black-Scholes approach to option-pricing. I conclude by introducing a new financial integration index which offers a number of benefits to existing approaches.