Short-term electricity load forecasting using time series and ensemble learning methods

Day-ahead electricity load forecasts are presented for the ISO-NE CA area. Four different methods are discussed and compared, namely seasonal autoregressive moving average (SARIMA), seasonal autoregressive moving average with exogenous variable (SARIMAX), random forests (RF) and gradient boosting regression trees (GBRT). The forecasting performance of each model was evaluated by two metrics, namely mean absolute percentage error (MAPE) and root mean square error (RMSE). The results of this study showed that GBRT model is superior to the others for 24 hours ahead forecasts. Based on this study we claim that gradient boosting regression trees can be appropriate for load forecasting applications and yield accurate results.