Forecasting Market Index Volatility Using Ross-Recovered Distributions

48 Pages Posted: 26 Oct 2018 Last revised: 31 Mar 2020

See all articles by Marie-Hélène Gagnon

Marie-Hélène Gagnon

Université Laval - Faculté d'Administration

Gabriel Power

Université Laval - Département de Finance et Assurance

Dominique Toupin

Bishop’s University - Williams School of Business

Date Written: June 1, 2019

Abstract

Ross (2015) shows that options data can reveal the market’s true expectations. Adapting this approach to index options (S&P, FTSE, CAC, SMI and DAX), we separate option-implied volatility into Ross-recovered true expected volatility and a risk preference factor. We investigate whether these factors perform better to forecast realized volatility if constructed locally or globally, yielding new insights to understand international dynamics in risk expectations and preferences. We find evidence of significantly improved realized volatility forecasts. Models using Ross-recovered value-weighted global measures of risk preferences have the best forecasting performances across indices. Risk preferences are best measured globally. Overall, the findings suggest that the Recovery Theorem is useful empirically and accurately recovers the true expected returns distribution and its associated pricing kernel.

Keywords: ross recovery; risk-neutral; volatility; forecast; options; international

JEL Classification: G12, G13, G14, G15

Suggested Citation

Gagnon, Marie-Hélène and Power, Gabriel and Toupin, Dominique, Forecasting Market Index Volatility Using Ross-Recovered Distributions (June 1, 2019). Available at SSRN: https://ssrn.com/abstract=3259654 or http://dx.doi.org/10.2139/ssrn.3259654

Marie-Hélène Gagnon

Université Laval - Faculté d'Administration ( email )

Quebec G1K 7P4
Canada

Gabriel Power (Contact Author)

Université Laval - Département de Finance et Assurance ( email )

Pavillon Palasis-Prince
Quebec G1K 7P4
Canada

Dominique Toupin

Bishop’s University - Williams School of Business ( email )

Canada

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
153
Abstract Views
871
rank
238,746
PlumX Metrics