--- On Tue, 12/1/10, Paul Seed <[email protected]> wrote:
> I have a puzzle. Is it better to use -xtreg, gls- or
> -xtreg, mle- for panel data with missing values? Little &
> Rubin have suggested that maximum likelihood estimation
> corrects for MCAR data if the model is correctly specificed.
I don't see why MCAR would be a problem at all (appart from
the smaller sample size): ignoring observations with missing
values if the missingness mechanism is MCAR just means you
have drawn a random sample from your random sample, so no
bias will be the result of that. So that would mean that if
you know your process is MCAR you can use whatever method
you like.
-- 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/
