Modelling the Joint Dynamics of Oil Prices and Investor Fear Gauge

Abstract

This paper investigates the interdependent relationship between WTI returns and the newly published crude oil volatility index (OVX), combining a cross-correlation function approach, a time-varying parameter (TVP) GARCH model, and a multivariate regression analysis, by which the direction, dynamics, magnitude and asymmetry of their relationship are modelled. At the same time, the implied volatility indexes in the stock market and the gold market are considered for comparison. It is found that there is a significant unidirectional causality-in-mean from WTI returns to the OVX changes, while causality-in-variance from the OVX changes to WTI returns is also significant. The contemporaneous relationship between the OVX changes and WTI returns is significantly negative, and their asymmetric relationship implies that OVX has played a greater role as a gauge of investor fear than as risk preference. The time-varying results indicate that the relationship between the changes in OVX and WTI returns is not always negative.

Keywords: OVX; Implied volatility index; Fear gauge; Time-varying relationship; TVP GARCH model