Skip to Content
EnergiesEnergies
  • Article
  • Open Access

26 December 2019

24 Pages

A Test of Using Markov-Switching GARCH Models in Oil and Natural Gas Trading

,
and
1
Faculty of Accounting and Management, Saint Nicholas and Hidalgo Michoacán State University (UMSNH), 58030 Morelia, Mexico
2
Financial Economy and Accounting Department, Faculty of Business, Finance and Tourism, University of Extremadura, 10071 Cáceres, Spain
*
Author to whom correspondence should be addressed.

Abstract

In this paper, we test the use of Markov-switching (MS) GARCH (MSGARCH) models for trading either oil or natural gas futures. Using weekly data from 7 January 1994 to 31 May 2019, we tested the next trading rule: to invest in the simulated commodity if the investor expects to be in the low-volatility regime at t + 1 or to otherwise hold the risk-free asset. Assumptions for our simulations included the following: (1) we assumed that the investors trade in a homogeneous (Gaussian or t-Student) two regime context and (2) the investor used a time-fixed, ARCH, or GARCH variance in each regime. Our results suggest that the use of the MS Gaussian model, with time-fixed variance, leads to the best performance in the oil market. For the case of natural gas, we found no benefit of using our trading rule against a buy-and-hold strategy in the three-month U.S. Treasury bills.

Article Metrics

Citations

Article Access Statistics

Multiple requests from the same IP address are counted as one view.