Risk management in the energy trading activity - an approach by using Multi Objective Genetic Algorithm and multi criteria theory

Abstract

In the new competitive environment of the electricity market, risk analysis is a powerful tool to guide investors under both contract uncertainties and energy prices of the spot market. Methodologies that allow what-if analysis involving simulation of spot price scenarios and energy contract's performance evaluation, are important to the decision maker and in particular to the trader, in order to foresee opportunities and possible threats in the energy market. The best solution to the problem of energy contracts portfolio optimization is a tradeoff solution, involving both risk and return, where the decision maker can know the worst scenarios of contracting and the spot market price. This paper proposes an approach for solving the contracts portfolio optimization problem by using Multi Objective Genetic Algorithm. In this approach, the risk metrics VaR and CVaR are considered as constraints in the construction of the Pareto Efficient Frontier.