R help - Apr 2012 - HoltWinters() fitted values

Hi everyone,

I'm using the HoltWinters() function to do a time series analysis.  The
function only returns the back fitted values ($fitted) after the first year
of data, which is my case, is a little more than half.  However, when I use
the plot() function, it plots the back fit for almost the entire data set.

Any ideas on how to extract the fitted values going all the way back to the
start of the data set?

Thanks,
Jess

--
View this message in context:
http://r.789695.n4.nabble.com/HoltWinters-fitted-values-tp4591103p4591103.html
Sent from the R help mailing list archive at Nabble.com.

HoltWinters doesn't give fitted values for the entire time series
since, like most time series models, there's a "recursive" element
to
it. I.e., if a very simple model is X_t = 1/2*(X_{t-1} + X_{t-2}) +
epsilon, you need to take X_0, X_1 as is to "start" the series.

If you run example(HoltWinters) you'll note that the red lines have
roughly one period's (I think that's the right term) less points
because those are needed to set up seasonality.

You can change the model to something simpler to get earlier fits, but
to the best of my understanding, you can't get them all the way to the
beginning with HoltWinters smoothing.

Michael

On Thu, Apr 26, 2012 at 5:57 PM, slipkid90515
<jessica.lyn.olsen at gmail.com> wrote:
> Hi everyone,
>
> I'm using the HoltWinters() function to do a time series analysis. ?The
> function only returns the back fitted values ($fitted) after the first year
> of data, which is my case, is a little more than half. ?However, when I use
> the plot() function, it plots the back fit for almost the entire data set.
>
> Any ideas on how to extract the fitted values going all the way back to the
> start of the data set?
>
> Thanks,
> Jess
>
> --
> View this message in context:
http://r.789695.n4.nabble.com/HoltWinters-fitted-values-tp4591103p4591103.html
> Sent from the R help mailing list archive at Nabble.com.
>
> ______________________________________________
> 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.

Yeah, I was pretty sure that it needed one period to begin estimating, but it
just seemed odd to me that the plot function was able to interpolate all the
way back to the beginning (well, nearly the beginning), but the fitted
values only began after the first period.

Thanks for your response!
Jess

--
View this message in context:
http://r.789695.n4.nabble.com/HoltWinters-fitted-values-tp4591103p4593025.html
Sent from the R help mailing list archive at Nabble.com.