> --- Jessica [iso-8859-1] �lschl�ger asked:
> I've got an unbalanced panel dataset covering 30 OECD-countries (i)
> over a period of 15 years (t).
>
> Running for example a fixed effects regression, I lose a lot of
> observations due to the missings in some of the variables In some
> cases I only have about 50 observations left.
>
> Is there a way to some kind of "save on observations" i.e. tell stata
> to use all of the observations available irrespective of the fact
> that some of the observations for a country i (panel) at time t are
> missing?
--- Nick Cox <[email protected]> answered:
> It's not clear what you want Stata to do here given missings: try
> harder?
>
> If there are missings, the only way to get more out is to put more
> in, e.g. impute values where missings would be otherwise. And then
> you need to discount your results accordingly.
One way to do what Nick suggests is to do multiple imputation, which in
Stata is implemented as -ice-, (within Stata type -ssc install ice-,
-ssc install mim-, -help ice-, and -help mim-). You might want to do
some reading before you get into that.
Royston, P. 2004. Multiple imputation of missing values. Stata Journal
4(3): 227�241.
Royston, P. 2005. Multiple imputation of missing values: update. Stata
Journal 5(2): 188�201.
Royston, P. 2005. Multiple imputation of missing values: Update of ice.
Stata Journal 5(4): 527-536.
Hope this helps,
Maarten
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
__________________________________________________________
Sent from Yahoo! Mail - a smarter inbox http://uk.mail.yahoo.com
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
