EcoSta 2017: Start Registration
View Submission - EcoSta2017
A0811
Title: Automatic ARIMA modeling and its application Authors:  Dedi Rosadi - Universitas Gadjah Mada (Indonesia) [presenting]
Abstract: In some application of time series modelling, it is necessary to obtain forecast of various types of data automatically and possibly, in real-time. For instances, to forecast large number of univariate series every day, or to do a real-time processing of the satellite data. Various automatic algorithms for modeling ARIMA models are available in the literature, where here we will discuss three methods in particular. One of the method is based on a combination between the best exponential smoothing model to obtain the forecast, together with state-space approach of the underlying model to obtain the prediction interval. The second method, which is more advanced method, is based on X-13-ARIMA-SEATS, the seasonal adjustment software by the US Census Bureau. Last method is a heuristic method based on genetic algorithm approach. These approaches are implemented in our R-GUI package RcmdrPlugin. Econometrics which now already integrated in our new and more comprehensive R-GUI package RcmdrPlugin.SPSS. We provide application of the methods and the tool using real data.