This work presents an empirical approach to modeling weekly Swedish electricity prices for 2005-2009. The main purpose is to obtain a model that enables to describe the behavior of the prices far in the future. A natural approach when modeling electricity prices is to describe the prices by an autoregressive process. Fell (2008) estimates the relationship between Nordic electricity prices, the EU emission trading scheme CO₂ allowance prices (EUA prices), and the prices of various generation fuels through a cointegrated vector autoregressive approach. Ferkingstad et al. (2010) study the price dynamics between Nordic and German electricity prices and major fuel sources (oil, gas and coal), and combine vector autoregression and vector error correction modeling with causal inference methods.

If the interest is in forecasting electricity prices far in the future, or investigating how different scenarios affect the prices, it is less appropriate to use lagged prices for predicting future prices. Therefore, we use regression techniques and model the mean electricity price with help of different exogeneous variables. It turns out that water reservoir contents in Sweden and Norway, inflow to the Swedish and Norwegian hydropower reservoirs, EUA prices, and electricity import and export to and from Sweden are essential factors for describing the variation in Swedish electricity prices. The estimated regression function is used to investigate what influence the explanatory variables in the model have on the Swedish area price. This is done by studying the behavior of the electricity price under different conceivable combinations of the explanatory variables. The regression residuals are described by a time series model.

References:

Fell, H. (2008). EU-ETS and Nordic electricity: a CVAR analysis, Resources for the Future Discussion, Paper No. 08-31.

http://www.rff.org/documents/RFF-DP-08-31.pdf

Ferkingstad, E., Løland, A., and Wilhelmsen, M. (2010). Causal modeling and inference for electricity markets. Energy Economics, http://dx.doi.org/10.1016/j.eneco.2010.10.006

Keywords: Swedish electricity prices; water reservoir content; inflow to water reservoirs; European Union Allowances

Biography: Kristi Kuljus took the bachelor degree in mathematical statistics at University of Tartu in Estonia. She continued her studies in Sweden and got her PhD in mathematical statistics at Uppsala University in 2008. She is currently working as a researcher at the Centre of Biostochastics at the Swedish University of Agricultural Sciences in Umeå.