Portfolio optimization under lower partial moments in emerging electricity markets: Evidence from Turkey

This paper presents a comprehensive comparison of mean-variance, down-side, and semi-variance methods for optimization in electricity markets and the corresponding methodologies to maximize the return while minimizing risk. Real Turkish day-ahead market data set between December 2009 and December 2012 is used in numerical calculations. Generation cost data of Hydraulic plants, lignite coal fired thermal power plants, and natural gas combined cycle power plants are taken into consideration in the course of optimization evaluations. In the present of real data, these methods can also be applied to renewable energy generation types. These three methods were able to be applied to all case scenarios effectively and produced efficient frontiers, optimal/minimal portfolios, and utility functions successfully. The results have revealed that the methods significantly provide decisions for power suppliers with different risk aversion levels, and for various instruments to maximize the profit while minimizing the associated market risks, and to meet generation obligations. Consequently, financial optimization under Lower Partial Moments constraints would give notable results in analyzing the efficient frontiers for electricity markets in Turkey.