Forecasting Wholesale Electricity Prices: A Review of Time Series Models

Weron, Rafal

Forecasting Wholesale Electricity Prices: A Review of Time Series Models

FINANCIAL MARKETS: PRINCIPLES OF MODELLING, FORECASTING AND DECISION-MAKING, W. Milo, P. Wdowinski, eds., FindEcon Monograph Series, WUL, Lodz, 2008

10 Pages Posted: 22 Jul 2008

See all articles by Rafal Weron

Rafal Weron

Wroclaw University of Science and Technology, Department of Operations Research

Date Written: July 22, 2008

Abstract

In this paper we assess the short-term forecasting power of different time series models in the electricity spot market. We calibrate autoregression (AR) models, including specifications with a fundamental (exogenous) variable - system load, to California Power Exchange (CalPX) system spot prices. Then we evaluate their point and interval forecasting performance in relatively calm and extremely volatile periods preceding the market crash in winter 2000/2001. In particular, we test which innovations distributions/processes - Gaussian, GARCH, heavy-tailed (NIG, alpha-stable) or non-parametric - lead to best predictions.

Keywords: Electricity price forecasting, heavy tailed distribution, autoregression model, GARCH model, non-parametric noise, system load

JEL Classification: C22, C46, C53, Q40

Suggested Citation: Suggested Citation

Weron, Rafal, Forecasting Wholesale Electricity Prices: A Review of Time Series Models (July 22, 2008). FINANCIAL MARKETS: PRINCIPLES OF MODELLING, FORECASTING AND DECISION-MAKING, W. Milo, P. Wdowinski, eds., FindEcon Monograph Series, WUL, Lodz, 2008, Available at SSRN: https://ssrn.com/abstract=1168382