Modelling Dynamic Conditional Correlations in Wti Oil Forward and Futures Returns

33 Pages Posted: 11 Jun 2004

See all articles by Matteo Manera

Matteo Manera

University of Milan-Bicocca, Italy - Department of Economics, Management and Statistics (DEMS); Fondazione Eni Enrico Mattei (FEEM), Milan, Italy

Alessandro Lanza

Fondazione Eni Enrico Mattei (FEEM), Milan; CMCC - Centro Euro-Mediterraneo sui Cambiamenti Climatici; Bocconi University - IEFE Centre for Research on Energy and Environmental Economics and Policy

Michael McAleer

Erasmus University Rotterdam - Erasmus School of Economics, Econometric Institute; Tinbergen Institute; University of Tokyo - Centre for International Research on the Japanese Economy (CIRJE), Faculty of Economics

Date Written: April 2004

Abstract

This paper estimates the dynamic conditional correlations in the returns on WTI oil one-month forward prices, and one-, three-, six-, and twelve-month futures prices, using recently developed multivariate conditional volatility models. The dynamic correlations enable a determination of whether the forward and various futures returns are substitutes or complements, which are crucial for deciding whether or not to hedge against unforeseen circumstances. The models are estimated using daily data on WTI oil forward and futures prices, and their associated returns, from 3 January 1985 to 16 January 2004. At the univariate level, the estimates are statistically significant, with the occasional asymmetric effect in which negative shocks have a greater impact on volatility than positive shocks. In all cases, both the short- and long-run persistence of shocks are statistically significant. Among the five returns, there are ten conditional correlations, with the highest estimate of constant conditional correlation being 0.975 between the volatilities of the three-month and six-month futures returns, and the lowest being 0.656 between the volatilities of the forward and twelve-month futures returns. The dynamic conditional correlations can vary dramatically, being negative in four of ten cases and being close to zero in another five cases. Only in the case of the dynamic volatilities of the three-month and six-month futures returns is the range of variation relatively narrow, namely (0.832, 0.996). Thus, in general, the dynamic volatilities in the returns in the WTI oil forward and future prices can be either independent or interdependent over time.

Keywords: Constant conditional correlations, Dynamic conditional correlations, Multivariate

JEL Classification: C32, G10, Q40

Suggested Citation

Manera, Matteo and Lanza, Alessandro and McAleer, Michael, Modelling Dynamic Conditional Correlations in Wti Oil Forward and Futures Returns (April 2004). FEEM Working Paper No. 72.04. Available at SSRN: https://ssrn.com/abstract=546484 or http://dx.doi.org/10.2139/ssrn.546484

Matteo Manera (Contact Author)

University of Milan-Bicocca, Italy - Department of Economics, Management and Statistics (DEMS) ( email )

Via Bicocca degli Arcimboldi, 8
Milan, 20126
Italy
+39 02 6448 5819 (Phone)
+39 02 6448 5878 (Fax)

HOME PAGE: http://www.matteomanera.it

Fondazione Eni Enrico Mattei (FEEM), Milan, Italy ( email )

Corso Magenta, 63
Milan, 20123
Italy
+39 02 520 36944 (Phone)

HOME PAGE: http://www.feem.it

Alessandro Lanza

Fondazione Eni Enrico Mattei (FEEM), Milan

Corso Magenta 63
20123 Milan
Italy

CMCC - Centro Euro-Mediterraneo sui Cambiamenti Climatici ( email )

via Augusto Imperatore, 16
Bologna, I-73100
Italy

Bocconi University - IEFE Centre for Research on Energy and Environmental Economics and Policy ( email )

viale Filippetti, 9
Milan, 20122
Italy

Michael McAleer

Erasmus University Rotterdam - Erasmus School of Economics, Econometric Institute ( email )

Rotterdam
Netherlands

Tinbergen Institute

Rotterdam
Netherlands

University of Tokyo - Centre for International Research on the Japanese Economy (CIRJE), Faculty of Economics

Tokyo
Japan

Register to save articles to
your library

Register

Paper statistics

Downloads
634
Abstract Views
6,560
rank
40,240
PlumX Metrics