aarjmd - Asian Academic Research
Transcription
aarjmd - Asian Academic Research
AARJMD VOLUME 3 ISSUE 2 (FEBRUARY 2016) ISSN : 2319 - 2801 A Peer Reviewed International Journal of Asian Academic Research Associates AARJMD ASIAN ACADEMIC RESEARCH JOURNAL OF MULTIDISCIPLINARY STATISTICAL FORECASTING USING BOX AND JENKINS APPROACH KHERBOUCHE MUSTAPHA1; SOUAR YOUCEF2 Abstract This study is designed to forecast the Subscriptions services prepaid in Mobilis exactly in saida, by using Box-Jenkins approach and to choose the best model of family models ARIMA. The data used for this paper was monthly data collected for a period of 2009-2012. The research found that the time series of subscriptions for the service prepaid in Mobilis is non-stationary and to make it stationary the first differences are applied. Differencing model were used to obtain the best model of family ARIMA. The most adequate model among the models that have been developed in this study for forecasting the number of subscriptions is an ARIMA model (2.1.0). Keys words: Forecast, times series analysis, (Box-Jenkins) Stationary and ARIMA models Asian Academic Research Journal of Multidisciplinary www.asianacademicresearch.org AARJMD VOLUME 3 ISSUE 2 (FEBRUARY 2016) ISSN : 2319 - 2801 References 1-Domadar N Gujariti économétrie, les édition de Boeck paris 1er Edition 2004. 2-George EP Box et G.Wilym M .jenkins , time and series analyses forcasting and control revised édition california 1976. 3-Regis Bourbonnais ,économetrie 8eme édition dunod , Paris 2011. 4-Regis Bourbonnais et Jean claude Usunier , prevision des ventes 3eme édition économie paris 2001. 5-Sami Khedhiri, cours d’économétrie méthodes et applications la voisier Paris 2007. 6-Valérie mignon, économétrie théorie et application , économica Paris 2008. Asian Academic Research Journal of Multidisciplinary www.asianacademicresearch.org