Researchers have shown that incorporating the intra-day relationships of electricity prices and trading volumes improves the accuracy of forecasts of daily electricity spot prices.
A power company able to forecast spot prices with a reasonable accuracy can adjust its own production or consumption schedule to maximize profits or reduce risk.
Katarzyna Maciejowska and Rafal Weron of the Hugo Steinhaus Center at the Wrocław University of Technology in Poland, used half-hourly data from the UK power market to model the spot prices directly (via ARX and Vector ARX models) and indirectly (via factor models).
The forecasting performance of five econometric models was evaluated and compared with that of a univariate model, which uses only (aggregated) daily data.
“The results indicate that there are forecast improvements from incorporating the disaggregated data, especially, when the forecast horizon exceeds one week”, the researchers say.
Additional improvements are achieved when the correlation structure of the intra-day relationships is explored.