Comparative Approach of Box-Jenkins Models and Artificial Neural Network Models on Births per Month in Gaza Strip Using R

Suhaila A. Baker, Bisher M. Iqelan


Comparative studies of different forecasting techniques can facilitate the selection of the best time series model for forecasting future expectations. In the present study, we address this problem by comparing the forecasting performance of the SARIMA model and four typical artificial neural networks, namely, MLP, ERNN, JRNN, and RBFNN in short-term forecasting for Births in Gaza Strip.

Analyses of Neural network models are done by the package (RSNNS) that is implemented in R program.

We conclude that forecasting with ANNs is accurate and more efficient than the SARIMA. In addition, the most accurate ANNs model among the four examined Neural Networks is RBFNN.


ARIMA, MLP, Elman Neural Network, Jordan Neural Network, RBF Neural Network, RSNNS Package.

Full Text:



  • There are currently no refbacks.

Copyright (c) 2017 IUG Journal of Natural Studies

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.