An integrated approach to optimize moving average rules in the EUA futures market based on particle swarm optimization and genetic algorithms

Climate change is a big challenge facing global community in 21st century. The carbon emission futures markets has been treated as a key tool to combat climate change cost-effectively. Making profits from futures trading is the fundamental incentive mechanism to keep this market run sustainably and effectively, while few technique analysis research on this topic has been done in the energy finance field. This paper contributes to the literature by proposing an integrated moving average rule for the European Union Allowance (EUA) futures market and designing an approach to optimize the weights of rules based on Particle Swarm Optimization (PSO) and Genetic Algorithms (GAs). The similarity of trading rules designed here is used to select base rules. An integrated approach based on PSO and GAs is proposed to identify the optimal weights group for the selected base rules. A group of Adaptive Moving Average trading rules with different weights constitutes an integrated trading rule. Experiments using the EUA futures market price were conducted. The results show that: (1) our model is profitable in the EUA future market with the proper parameter except the case that prices fluctuate significantly; (2) the adjustment cycle of 5days is more useful than 20days or 50days; (3) the algorithm achieves the best performance at the 0.78 similarity threshold; (4) the rule with the short period of 150days and the long period of 200days is a useful building block for a successive rule set. This approach is a useful reference to the practical investments in EUA futures market.