Structural Stochastic Volatility
(with Federico Bandi and Roberto Reno')

We present a new methodology for estimating spot equity characteristics (i.e. spot volatility, sport leverage, and spot volatility-of-volatility) from option data. Within a simple structural model we show both empirically and theoretically a strong link between financial leverage and the three equity characteristics.
Stacks Image 1109

Abstract

We use a local (in time) expansion of the characteristic function of the equity process in continuous time to derive short-maturity option prices. The prices, along with data on short- maturity options, are employed to jointly identify equity characteristics (spot volatility, spot leverage and spot volatility of volatility) which have been the focus of separate strands of the literature. We show that the proposed identification method yields measurements which are statistically accurate and economically revealing. Interpreting equity as a call option on asset values, all equity characteristics should depend on fundamental state variables, such as the vari- ance of the firm’s assets and the extent of the firm’s financial leverage. Among other findings, consistent with economic logic, we document a strong link between spot leverage (the generally- negative correlation between equity returns and spot volatility) and financial leverage (the firm’s debt-to-equity ratio), a relation invariably found to be elusive in the data. We conclude that the economic content of option-implied measurements can be put to work to study the structural drivers of equity (and debt) return dynamics from a novel vantage point.



Testing for Asset Price Bubbles using Options Data
(with Robert Jarrow and Sujan Lamichhane)

We show that option data can be effectively utilized to detect and quantify bubbles in the underlying asset.
Stacks Image 1119
We present a new approach to identifying asset price bubbles based on options data. We estimate asset bubbles by exploiting the differential pricing between put and call options. We apply our methodology to two stock market indexes, the S&P 500 and the Nasdaq-100, and two technology stocks, Amazon and Facebook, over the 2014-2018 sample period. We find that, while indexes do not exhibit significant bubbles, Amazon and Facebook show frequent and significant bubbles. The estimated bubbles tend to be associated with high trading volume and earning announcement days. Since our approach can be implemented in real time, it is useful to both policy-makers and investors. As an illustration we consider two case studies: the Nasdaq dot-com bubble (between 1999 to 2002) and GameStop (between December 2020 and January 2021). In both cases we identify significant and persistent bubbles.