High and Low Intraday Commodity Prices. A Fractional Integration and Cointegration approach

This paper examines the behaviour of high and low prices of four commodities, namely crude oil, natural gas, gold and silver, and of the corresponding ranges using both daily and intraday data at various frequencies. For this purpose, it applies fractional integration and cointegration techniques; in particular, an FCVAR model is estimated to capture both the long-run equilibrium relationships between high and low commodity prices, referred to as the range, and the long-memory properties of their linear combination. Fractional cointegration in found in all cases, with the range showing stationary and nonstationary patterns and changing substantially across the frequencies. The findings may assist investors in improving their trading strategies since high and low prices serve as entry and exit signals in the market.