Applying modern portfolio theory for a dynamic energy portfolio allocation in electricity markets

• Published in: Electric power systems research 2017-09, Vol.150, p.11-23
• Main Authors: Garcia, Reinaldo C., González, Virginia, Contreras, Javier, Custodio, Janiele E.S.C.
• Format: Article
• Language: English
• Subjects: Conditional Value at Risk; Deregulation; Electric power; Electricity; Electricity distribution; Electricity generation; Electricity market; Markets; Mean variance; Optimization; Portfolio theory; Pricing
• ISSN: 0378-7796; 1873-2046
• DOI: 10.1016/j.epsr.2017.04.026

•This paper proposes two portfolio models, one applying the Mean Variance Criterion (MVC) and the other one the Conditional Value at Risk (CVaR), applied to electricity markets.•The MPT models are combined with a generalized autoregressive conditional heteroskedastic (GARCH) prediction technique for a Genco to optimally diversify their energy portfolio.•The two models are applied to the PJM electricity market. In deregulated electricity markets, a Generation Company (Genco) has to optimally allocate their energy among different markets including spot, local and bilateral contract markets. Modern portfolio theory (MPT) allows a Genco to achieve their goal by maximizing their profit and decreasing their associated risk. Combining MPT with an adequate tool to forecast energy prices makes it possible for a Genco to vary the optimal allocation of their portfolio even on a daily basis. This paper proposes two MPT models, one applying the Mean Variance Criterion (MVC) and the other one the Conditional Value at Risk (CVaR). The MPT models are combined with a generalized autoregressive conditional heteroskedastic (GARCH) prediction technique for a Genco to optimally diversify their energy portfolio. The two models are applied to a real PJM electricity market showing not only their capabilities but also useful comparisons between them in order to help decision makers to use them as decision-aid tools.