Is the VIX a Reliable Indicator of Stock Market Volatility?
MetadataShow full item record
This thesis examines the reliability of the Chicago Board Options Exchange Volatility Index (VIX) as an indicator of realized stock market volatility. The VIX is published daily by the Chicago Board Options Exchange (CBOE) and is widely referred to as the ‘fear gauge’ in the market. The VIX is computed from the relevant S&P 500 call options prices, put options prices and the forward S&P 500 index. The VIX is our measure of implied volatility in this thesis. We proxy realized market volatility using the daily range of the S&P 500 index. Using a GARCH (1,1) model, we find a positive, statistically significant, contemporaneous relationship between the daily closing values of the VIX and the range of the market index. Our results also support a stronger magnitude of relationship between increases in the value of the VIX and realized market volatility, compared to the relationship between decreases in the value of the VIX and realized market volatility. This finding confirms the VIX as a reliable measure of market volatility for market participants. Additionally, in a major contribution to existing literature, we separately model changes in the VIX that occur during non-trading hours vis-à-vis changes in the VIX that occur during trading hours. We find that changes in the VIX during non-trading hours predict realized market volatility; up to four days. Additionally, changes in the VIX during trading hours predict realized market volatility; up to five days. Notably, the magnitude of this relationship is still positive and mostly diminishes as the time lag increases. Our analyses of the separate changes in the VIX during non-trading hours and changes in the VIX during trading hours highlight the need for market participants to isolate changes in the VIX that occur during these two different times. These isolated changes in the VIX are more informative for predicting realized volatility compared to analyses that only consider changes in the closing values of the VIX. Furthermore, we find that the scale of the relationship between the VIX and realized market volatility is strongest on Fridays compared to other trading days of the week. This increase in the magnitude of the relationship between the VIX and realized volatility around the ‘weekend’ is likely due to uncertainties that are associated with news including major changes in economic or monetary policy, company earnings etc. that are typically released around the end of the week. These uncertainties are reasonably captured in the VIX. In sum, the evidence provided by our thesis supports the VIX as a reliable tool for predicting realized market volatility for up to five days ahead, with positive changes in the VIX on Fridays being very informative.
DegreeMaster of Science (M.Sc.)
SupervisorTannous, George; Yang, Fan
Copyright DateDecember 2018
VIX, volatility, range, volatility index