R help - Apr 2009 - 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

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.