Applications of UNI Variate Time Series Models and Neural Networks

Forecasting of daily and monthly electricity load using Box-Jenkins methodology and feed forward neural networks is discussed. This study investigates application of neural networks models and the results of neural networks determination be compared with those obtained by Box-Jenkins method. The performances were compared based on three measures: mean absolute error (MAE) mean absolute percentage error (MAPE) and root mean square error (RMSE). The Final conclusion of this book is Feed-Forward neural networks models are better and superior than Box-Jenkins models.