Re: st: Disabling Kalman filter in arima

[Maarten buis <maartenbuis@yahoo.co.uk>]

--- On Thu, 16/9/10, kristian@hefting.dk wrote:
> Stata uses a Kalman filter to replace missing/unobserved
> data when fitting an arima model. Is there any way to
> disable this filter so the missing/unobserved data is
> treated as zeros?

As you can see in the example below -arima- just drops
observations with missing values, rather than replace
their values (look at the number of observations used). 
That is the default for (almost) all Stata programs.

*--------- begin example -----------
webuse wpi1, clear
arima wpi, arima(1,1,1)
replace wpi = . in 10
arima wpi, arima(1,1,1)
*--------- end example -------------

Replacing missing values manually with a fixed value is
usualy very bad practice, unless you have information that
a missing value is just an odd way that your data collectors
used for storing a 0. In that case you can just correct 
your data by typing:

replace <variable_name> = 0 if <variable_name> == .

Otherwise I would just stick with the Stata default.

Hope this helps,
Maarten

--------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany

http://www.maartenbuis.nl
--------------------------


      

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/