|
|
 |
| |
|
|
|
Comparison of ARIMA, ANN and Hybrid ARIMA-ANN Models for Time Series Forecasting |
|
|
|
PP: 1003-1016 |
|
|
|
doi:10.18576/isl/120238
|
|
|
|
Author(s) |
|
|
|
Amjad A. Alsuwaylimi,
|
|
|
|
Abstract |
|
|
| This paper aims to compare between Auto Regressive Integrated Moving Average (ARIMA) model, Artificial Neural Networks (ANN) and hybrid models for time series forecasting. The dataset used on this study is based on the monthly gold prices during Nov-1989 to Dec-2019. This dataset was used to train and test the predictive models. The performances were evaluated based on three metrics Mean Square Error (MSE), Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE) to determine the more appropriate model and evaluate models’ performance. The most important finding was that applying hybrid models can improve the forecasting accuracy over the ARIMA and ANN models. This may suggest that neither ARIMA nor ANN model captures all of patterns in the data.
|
|
|
|
|
|
|
|
