Key points are not available for this paper at this time.
In asset pricing applications one of the most powerful and widely used models is the famous Black-Sholes model. In applications of stock option pricing, the resulting Black-Scholes formula requires inputting several values, including the current price of the stock and the market volatility. The question of how to input the volatility in real-time can be very challenging, as commonly utilized measures, such as the volatility index (VIX), can be viewed as lagging economic indicators. This research discusses a scheme that applies a mathematical model which defines how far the current market value is above or below from where macroeconomic conditions would expect it to be. In prior research, a regression model was created to define a framework that theoretically outlined a new measure related to market volatility. When computing various financial predictions, such as evaluating the fair price of an option, this measure can be used to supplement common current measures, such as the VIX. In the current work, several mathematical methods, including machine learning, are investigated to determine if improvements to accuracy can be made, and define a practically usable scheme. Slight improvements to accuracy were discovered in comparison to the prior regression model; the Principal Component Analysis method is recommended for usage in real time applications.
Smith et al. (Tue,) studied this question.