Hi! I am exploring the forecast package (http://www.robjhyndman.com/ index.php?option=com_content&task=view&id=55&Itemid=71). I am doing ARIMA modelling with auto.arima() function. Is it possible to get the decomposition of a time series using the model found by auto.arima()? I would like to decompose a time series in trend, seasonal and irregular components according to the model. I have looked at the decompose() and stl() functions. But these function do not take into account the specific model found by auto.arima(). Thanks! Martins
Decomposition of time series with forecast package
2 messages · djhurio, Christofer Bogaso
1 day later
If I am right then you must get the seasonal factor etc (if any) out before fitting ant ARIMA (or statistical model) i.e. fit ARIMA on residual series not original series. -----Original Message----- From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On Behalf Of djhurio Sent: 14 April 2009 17:41 To: r-help at r-project.org Subject: [R] Decomposition of time series with forecast package Hi! I am exploring the forecast package (http://www.robjhyndman.com/ index.php?option=com_content&task=view&id=55&Itemid=71). I am doing ARIMA modelling with auto.arima() function. Is it possible to get the decomposition of a time series using the model found by auto.arima()? I would like to decompose a time series in trend, seasonal and irregular components according to the model. I have looked at the decompose() and stl() functions. But these function do not take into account the specific model found by auto.arima(). Thanks! Martins ______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.