In this thesis I have investigated a small part of the extensive Nordic electricity market. My intention was to model the forward Nordic “year contract” for 2010, and its dynamics in relation to a set of explanatory variables observed over three years. My goal was to derive interpretable and statistically significant results, allowing me to better understand the fundamentals in the above mentioned market.
The thesis is also an empirical and statistical investigation of certain truths regarding the relationship between variables in the Nordic electricity market. They are often referred to by market commentators and others who might have a say. An example of such a truth is the common assumption that the price of oil is a good indicator for the price of electricity. A notion I also find empirical evidence for in chapter six.
The findings presented in this thesis support most of the assumptions made in the above context, and can even suggest further explanation. However, they are based on data sets from a limited time period, which has to be taken into account.
The chapters are organized in the following manner; Chapter 2 is an introduction into the Nordic electricity market with a short summary of the most important aspects the reader needs to know to understand the following chapters. Chapter 3 presents the main hypothesis and assumptions regarding the variables included in the thesis and their function. It also holds demarcations and further specification of what is to be included in the thesis. Chapter 4.1-2 consists of the theoretical framework needed to understand how the estimation method has been conducted. 4.3-5 presents an exposition of the statistical tests and misspecification tests vital for statistical analysis. Lastly in 4.6, the use of the Autometrics algorithm has been accounted for. Chapter 5 presents the data sets that I have used, how they are denoted and what values they are measured in. Chapter 6.1-6 holds all the empirical results extracted from Models – 1 to 4, spanning from year 2007-2009. Chapter 7 is an extension of chapter 6, where I have used instrumental variables and two stage least squares to further investigate Model – 3 with observations from 2007. Chapter 8 concludes the thesis and suggests some interesting extensions for further investigation.