Predicting the subscribers of fixed-line and cellular phone in Japan by a novel prediction model

• We use two approaches to develop a novel prediction model.
• The cubic spline function is first applied to improve the prediction capability.
• Furthermore, we use particle swarm optimization algorithm to optimize proposed model.
• Prediction result for the amount of phone subscribers is satisfactory.

In this paper, a novel grey prediction model is proposed to enhance the performance of prediction for the amount of fixed-line and cellular phone subscribers in Japan. The cubic spline function is first integrated into grey prediction model to enhance its prediction capability. Then the particle swarm optimization (PSO) algorithm is applied, so that the prediction performance can be improved further. The prediction results using proposed models are very satisfactory.