Nonlinear relationships amongst the implied volatilities of crude oil and precious metals

Unlike most prior studies, we rely on implied volatility data from the Chicago Board of Options Exchange to investigate the presence of cointegration and nonlinear causality between the global market of crude oil and the markets of precious metal (gold and silver) and gold-miner stocks. We apply nonlinear ARDL bound and symmetric and asymmetric nonlinear Granger causality tests to examine cointegration and short-run "lead-lag" relationships, respectively. For comparison purposes, we also apply the linear ARDL models. The results show that the nonlinear ARDL models successfully capture the long-term linkages between oil and precious metal markets, while the linear ARDL approach mostly fails to do so. Results from the nonlinear Granger causality test suggest a bidirectional and symmetric effect between crude oil and gold markets. Investment and policy implications are discussed.