volatility index is reflective of the increasing uncertainty that precedes monetary policy
decisions about the Fed Fund Target rate. Similar evidence is obtained from alternative
markets including the results by Neely (2005) which indicate that changes in implied
volatility from Eurodollar interest rates coincide with major events related to monetary
policy decisions, real economy and equity markets. Also, Fornari (2004) suggests that
implied volatility from swaptions markets is reflective of market participants’ reaction to
macroeconomic announcements about the release of US economic indicators.
This close connection between monetary policy decisions and changes in implied
volatility adds to the early evidence on the informational content of implied volatility, which
is rather mixed. Indeed, the empirical studies by Day and Lewis (1992) based on S&P 100
index options, and by Lamoureux and Lastrapes (1993) based on individual stock options,
suggest that implied volatility is a biased and inefficient estimate of market volatility.
Canina and Figlewski (1993) provide also evidence of insignificant correlation between
S&P 100 implied volatility and future market volatility. However, other empirical results are
more supportive of the informational content of implied volatility. For instance, Harvey and
Whaley (1992) suggest that implied volatility provides an efficient forecast of future
volatility. The empirical evidence from Fleming (1998) also suggests that despite the
upward bias, the forecast errors associated with S&P 100 implied volatility are orthogonal
to parameters included in ARCH models. Moreover, Christensen and Prabhala (1998) find
that the forecasting performance of implied volatility is higher than historical volatility.
Furthermore, Blair, Poon and Taylor (2001) find that the S&P 100 implied volatility index is
more accurate for out-of-sample forecasting than realized volatility, irrespective of data
frequency and the forecasting horizon. The empirical tests by Becker, Clemens and White
(2006) suggest that the S&P 500 implied volatility index is efficient with respect to some but
not all the information set available for forecasting purposes. More recently, there is