An Econometric Analysis of Global Commodity Prices

George, Rhea (2019) An Econometric Analysis of Global Commodity Prices. PhD thesis, UNSPECIFIED.

[thumbnail of 2019rheaphd]
Text (2019rheaphd)
Restricted to Repository staff only until 17 February 2025.

Download (4MB)


Between 2002 and mid-2012, the Dow Jones-UBS commodity index of spot commodity prices increased by around 360%. Crude oil prices increased by a massive 138% between 2006 and 2008. Besides the magnitude of the price rise, what was remarkable was that the prices of a broad range of commodities: agricultural, energy and metal all went up and declined together. This highly volatile period also coincided with a huge influx of investments in the commodity markets and the emergence of commodities as an alternative investment class; a phenomenon popularly termed the financialization of commodities. Our study covers the period from 2004 to 2016. This avoids any effects from the dot-com crisis on our data. In this thesis, we study the evolution of commodity prices, and the increase in comovements across commodities as well as between commodities and the equity market. We extend our analysis by testing for the existence of contagion in the commodity markets and between commodities and the equity market. Our study covers eight commodities, namely: WTI (crude oil), gold, copper, wheat, soybeans, cocoa, rubber and wool, including all major commodity groups and the global benchmark for equity markets, the S&P 500 index. During the global financial crisis period, concurrent with the rapidly increasing index investments in commodities, we find strong comovements between WTI and the rest of the commodities. We also find comovements between the S&P 500 index and all the commodities except gold and cocoa, signifying financialization of the commodity markets. Further, we extract conditional variances and conditional correlations from a DCC-GJR-GARCH model. While testing for contagion we focus on two main crisis periods: the global financial crisis (GFC) and the European sovereign debt crisis (ESDC). Our results show that during the GFC period, there is evidence of contagion between oil markets and the soybeans and rubber markets as well as between the S&P 500 index and the wheat, rubber and wool markets. And during the ESDC period, contagion existed between the S&P 500 index and wheat, soybeans, cocoa and rubber markets. Clearly, the ESDC was more contagious than the GFC period. Throughout our study, we find multiple results that prove that gold remained insulated from other commodities and also from the equity market confirming its safe-haven properties and diversification benefits. Our results provide significant implications for international investors and policymakers. Against this background, we then carry out a study to investigate long memory properties and dynamic conditional correlations between the commodity futures markets and equity markets covering both the emerging market equities and the developed market equities. This study covered five commodity futures: WTI, gold, copper, wheat and cocoa, five equity markets from developed economies: USA, Germany, France, UK and Japan and the BRICS equities: Brazil, Russia, India, China and South Africa. To incorporate long term dependence and to estimate the dynamic conditional correlations, we use a DCC-FIGARCH model. We confirm the presence of significant long memory in all the commodities and equity markets in our sample and identify time-varying dynamic conditional correlations. We then construct and investigate optimal portfolio weights and dynamic hedge ratios for commodity-equity portfolios to illustrate the significance of hedging using commodities. We carry out a comparative analysis to differentiate portfolios with commodities and BRICS equities and those with commodities and developed market equities. We find that comovements are stronger between commodities and the BRICS equities and in addition, the hedge ratios show evidence of commodities being a cheaper and effective hedge when used with developed market equities. We conclude that commodities performed as better hedges in portfolios with developed markets equities.

Item Type:
Thesis (PhD)
ID Code:
Deposited By:
Deposited On:
17 Feb 2020 09:25
Last Modified:
18 Sep 2023 02:20