Title: Using conditional asymmetry to predict commodity futures prices

Authors: Fabio S. Dias

Addresses: Department of Statistical Science, University College London, Gower St, London WC1E 6BT, UK; Stalwart Holdings UK LLP, 61 Bridge Street, Kington HR5 3DJ, UK

Abstract: Despite decades of studies, there is still no consensus on what type of serial dependence, if any, might be present in risky asset returns. This manuscript provides an empirical study of the prices of energy commodities, gold and copper in the futures markets and demonstrates that, for these assets, the level of asymmetry of asset returns varies through time and can be forecast using past returns. A regime switching model is used to construct a managed futures trading strategy that provides returns that are statistically significant. It is also demonstrated how such model can be used to make probabilistic predictions of commodity prices in futures markets, which can be used to drive value-at-risk and potential future exposure metrics or guide dynamic hedging strategies of commodity price risk.

Keywords: time series analysis; time series momentum; probabilistic forecasting; mixture models.

DOI: 10.1504/IJFMD.2021.115876

International Journal of Financial Markets and Derivatives, 2021 Vol.8 No.2, pp.185 - 203

Received: 22 Apr 2020
Accepted: 11 Apr 2021
Published online: 25 Jun 2021 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article