Price forecasting in the day-ahead electricity market

Electricity markets are becoming more sophisticated and price forecasting is gaining importance for market participants to adjust their bids in the day ahead electricity market. Compared with load, the price of electricity is volatile, but it is not regarded as random; therefore it is possible to identify patterns based on historical data and load forecast. Market participants only have access to the public information provided by the independent system operator (ISO); this limits the ability of the market participants to produce accurate price forecast. An accurate estimation of the electricity price helps the market participants to determine their bidding strategy; making necessary the development of accurate price forecasting methods. In this paper the authors propose a statistical method for price forecasting considering the historical price and loads of the California power market; matching the load and price profiles of the forecasting day with historical data of previous similar days. Simulation results show that the forecasted price is closely related with the price historical data and the load of the forecasting day. The accuracy of the method was evaluated by comparing the actual price with the forecasted price and calculating the mean absolute percentage error (MAPE). The simulation results show the accuracy of the method and its practical application. The proposed method can help market participants to evaluate the risks associated with price volatility and helps them to determine their bidding strategy.