Electricity Price Modeling and Analysis

4 Introduction The energy markets, and especially the market for electricity, represent a subject of great interest. In particular, issues related to electricity price modeling and the relationships between electricity prices and other economic variables have a great potential for future progress of the academic research. This work is divided into three parts, each one dealing with a particular theme. Part One briefly describes the mechanics of the electricity market, with a particular focus on the organized electricity exchanges and on the description of derivatives contracts having electricity price as the underlying asset. In Part Two, after having assessed the most important features of electricity prices (namely, mean regression, spikes, time varying volatility and seasonality), we will try to pursue the ambitious goal of writing a proper model for the electricity spot price, which takes into account all their above qualities. We will start from a very basic random walk model, and step by step we will add more variables and parameters to deal with each of the typical features of electricity prices. Each model will be estimated using the maximum likelihood method, which is particularly indicated when dealing with energy commodities. The final step is a regime-switching model with two regimes, and, in order to evaluate its quality, we will perform some trajectorial and statistical analysis on the fitted model versus the actual data. In Part Three, starting from the model of Prescott (1986) and its definition of price trend, we will try to describe the cyclical behavior of electricity prices with respect to other economic and financial time series. Similar studies have been conducted with gas and oil prices, but only few authors treated the cyclicality of electricity prices in a specific way. The goal of this final part is to represent a starting point for future research.