Portfolio Optimization of Power Generation Assets

In this chapter I provide an overview of the theoretical and applied literature dealing with mean-variance portfolio analysis used to study the efficiency of portfolios of power generation assets. The relevant literature focuses on the risk-mitigating benefits of technological diversification vis-a-vis single-technology analysis with conventional levelized cost analysis, to varying degrees taking into account real-world constraints. Part of the cutting-edge research deals with the benefits that accrue from intra-technology diversification and geographical dispersion. Some studies also take into account country-specific differences in national regulatory framework conditions and local resource potentials (esp. in the case of wind power). Other research has focused on technical and system-related aspects, such as load dispatch and portfolio restrictions, e.g., arising from grid constraints and the intermittent nature of many renewable energy sources. Complementary approaches to mean-variance portfolio analysis, such as real options analysis and fuzzy modeling, as well as alternative measures of risk (e.g. Value at Risk – VaR, Conditional Value at Risk – CVaR, and (semi-) mean absolute deviation), are briefly discussed as well, thus acknowledging some of the most important recent developments in this research area that have not been reviewed elsewhere yet.