The goal of this paper is to investigate the factors that influence on co-volatility of commodities. For this purpose we conduct a two steps analysis. First, we use a Dynamic Conditional Correlation GARCH (DCC-GARCH) model of Engle (2002) to examine the time-varying conditional correlation between returns of ten commodities, including agriculture, metal and energy commodities, using daily data from 2008 to 2014. The finding of this step confirms the existence of conditional correlation or co-volatility between those markets. Then, in the second step we demonstrate an analysis on determinant factors behind those co-volatilities, using ARDL approach of Pesaran et al. (1999). The determinant factors which we examine are categorized under three groups, including a) macroeconomics fundamentals, b) physical supply and demand fundamentals, proxied by inventory level, using time varying co-movements of markets inventories and c) market fundamentals, applying time varying co-movements of markets speculations as well as markets liquidities.

The results indicate that among the explanatory factors, the ones with the strongest effects are business cycles, proxied by ADS and the US exchange rate, which both negatively influence on co-volatility of the majority of commodities, also the uncertainty in stock market, proxied by VIX, shows a strong positive influence on markets co-volatility. Moreover, there are evidences supporting that co-movements of markets inventories, markets speculations and markets liquidities influence on markets co-volatility, however, their effects are not as strong as the effects of macroeconomic factors and the signs of their effects vary among the markets.