Beschreibung:
The spillover effect is a significant factor impacting the volatility of commodity prices. Unlike earlier studies, this research uses a relatively recent estimation technique such as rolling window-based Quantile VAR (QVAR) advanced by Ando et al. (2018) to describe the conditional volatility spillover and connectedness between energy, biofuel and agricultural commodity markets in a more realistic way. Since the magnitude of connectedness and spillover effects may switch between bearish and bullish markets states over time, a QVAR is a relatively realistic and appropriate approach to capture the connectedness as compared to the mean-based approaches of Diebold and Yilmaz (DY; 2009, 2012, & 2014) which are mostly used in the literature. To this end, we employ volatility estimates by using the realized variance advanced by Parkinson (1980). Specifically, we investigate the time-varying volatility spillovers and connectedness between agricultural (wheat, corn, sugar, soyabean, coffee, cotton), energy (gasoline, crude oil, natural gas) and biofuel (ethanol) commodities from January 12, 2012 to May 10, 2021 using the QVAR model. By comparing our empirical analysis with results from the DY spillover model, we establish that connectedness is stronger in the left and right quantiles than those in the mean and median of the conditional distribution, emphasizing the importance of systematic risk spillovers during extreme market movements. Furthermore, we find that volatility spillovers and connectedness in the right tail is higher than in the left tail. In particular, we document significant volatility spillovers from agricultural commodities to energy markets during extreme markets conditions and observe the dominance of agricultural commodities over energy markets. Finally, some useful implications are summarized for investors' portfolios and risk avoidance