R help - Feb 2011 - Estimation and Forecast of Seasonal Component

Hi list,


I would like to estimate and forecast the seasonal component of a series. My
model which uses daily data would be something y t = alpha + beta x SeasComp
t + gamma x OtherRegressors t.

One approach to this would be use quarterly dummies, another to use a sine
function. The first would cause a step change when we move from a season to
another;  the latter impose too much regularity in my data, i.e. symmetry
around y-axis and x-axis.

I  have looked at decompose and stl but could  not find a way to forecast
the seasonal component. They don’t allow for they presence of other
regressors either. Can someone let me know the best way for me to proceed
with this?
Thanks in advance
Paolo

	[[alternative HTML version deleted]]

I would probably use a Dynamic Linear Model, combining seasonal
component and regression. See package dlm and its vignette for
examples. 

HTH,
Giovanni

On Tue, 2011-02-01 at 11:01 +0000, Paolo Rossi wrote:
> Hi list,
> 
> 
> I would like to estimate and forecast the seasonal component of a series.
My
> model which uses daily data would be something y t = alpha + beta x
SeasComp
> t + gamma x OtherRegressors t.
> 
> One approach to this would be use quarterly dummies, another to use a sine
> function. The first would cause a step change when we move from a season to
> another;  the latter impose too much regularity in my data, i.e. symmetry
> around y-axis and x-axis.
> 
> I  have looked at decompose and stl but could  not find a way to forecast
> the seasonal component. They dont allow for they presence of other
> regressors either. Can someone let me know the best way for me to proceed
> with this?
> Thanks in advance
> Paolo
> 
> 	[[alternative HTML version deleted]]
> 
> ______________________________________________
> 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.