Forecasting telecommunications data with Autoregressive Integrated Moving Average models

Abstract

Forecasting of telecommunication data find difficult according to International Telecommunication Union (ITU) due to uncertainty involved and the continuous growth of data in telecommunication markets as it helps them in planning and determining their networks. So, there is a need of good forecasting model to predict the future. In this paper, Autoregressive Integrated Moving Average model is utilized for forecasting telecommunication data. This model adaptively uses auto regression, moving average or combined together if required. The major steps involved in the ARIMA model is identification, estimation and forecasting. The adaptive ARIMA model is then applied to M3-Competition Data to do forecasting of telecommunication data. The performance of the model is found out using the evaluation metrics such as Sum of Squared Regression, Root Mean Square Error, Mean Absolute Deviation, Mean Absolute Percentage Error and Maximum Absolute Error. The results proved that the ARIMA models provide 7.6% improvement than the neural network method in forecasting performance.