Title :
The Application of ARIMA Model in Chinese Mobile User Prediction
Author_Institution :
Marketing Dept., Alcatel-Lucent Shanghai Bell, Shanghai, China
Abstract :
This paper proposes another technique to forecast Chinese mobile user based on the ARIMA model. There are two major premises to use ARIMA model in mobile user prediction: mobile market is mature and still has big space for development. On the basis of the historical incremental mobile user number from 2002q1 to 2009q4, through model identification and estimation, an ARIMA (1, 1, 2) model is selected to fit the mobile user time series. After the model diagnostic evaluation, the 2010 quarterly incremental mobile user number is predicted. The model is approved to be accurate for short-time mobile user prediction.
Keywords :
autoregressive moving average processes; forecasting theory; identification; market research; mobile communication; prediction theory; telecommunication industry; time series; ARIMA model; Chinese mobile user forecasting; Chinese mobile user prediction; mobile market; model diagnostic evaluation; model estimation; model identification; time series; Biological system modeling; Correlation; Estimation; Industries; Mobile communication; Predictive models; Telecommunications; ARIMA model; Mobile user prediction;
Conference_Titel :
Granular Computing (GrC), 2010 IEEE International Conference on
Conference_Location :
San Jose, CA
Print_ISBN :
978-1-4244-7964-1
DOI :
10.1109/GrC.2010.31
