Big Data and Machine Learning in Quantitative Investment

CHAPTER 9Machine Learning and Event Detection for Trading Energy Futures

Peter Hafez and Francesco Lautizi

9.1 INTRODUCTION

The commodity futures spectrum is an integral part of today's financial markets. Specifically, energy‐related ones like crude oil, gasoline and natural gas, among many more, all react to the ebbs and flows of supply and demand. These commodities play a crucial role in everyday life, as they fuel most of the world's transportation systems and they are the input to businesses across all the industrial sectors, hence they are inherently linked to the economic cycle. Economic indicators such as gross domestic product and the unemployment rate to political upheaval and natural disasters, not to mention commodity‐specific issues like oil and gas pipeline disruptions or embargos, all contribute to the pricing of commodity futures (Table 9.1).

Table 9.1 Performance statistics

Source: RavenPack, January 2018.

Statistics	Out‐of‐sample
	Ensemble	High‐vol	Low‐vol
Annualized return	9.8%	21.3%	−3.0%
Annualized volatility	15.0%	16.9%	15.3%
Information ratio	0.65	1.27	−0.20
Hit ratio	51.1%	53.9%	47.5%
Max drawdown	38.3%	18.0%	62.2%
Per‐trade return (bps)	3.88	8.82	−1.97
Number of trades	2740	1929	811

The high‐vol and low‐vol strategies trade only during these regimes while the ensemble strategy trades irrespective of the regime. The out‐of‐sample period is January 2015 to December 2017.

In previous research, Brandt and Gao (2016) took a novel approach ...