Hello,
I tried to use the seemingly unrelated regression
command "suest" after running two xtnbreg models (both
random effects, and with approximately the same number
of observations). After estimating each equation (see
sample code below), I tried to include the "score"
option, which was not allowed by STATA ("option
score() not allowed").
Going through the relevant program menus from the
Cross-sectional time series, I entered the commands
for xtnbreg, selected the "Reporting" tab in the
xtnbreg menus. There is an option called "New
Generating Variables", followed by a grey area
(impossible to input text) that asks to name the
"Score variable." So, seems to be impossible to use
this option, eventhough the Stata help pages on
xtnbreg post-estimation commands and the suest command
indicate that it should be possible. Has anyone
encountered this? Any ideas to move forward?
Thanks,
Hakan Ener
France
Here is the detailed code I tried:
This was not allowed (error says "option score() not
allowed"):
xtnbreg Count a b c, re exposure(d) score(V)
Instead, I did:
Count a b c, re exposure(d)
followed by
predict z,score
which was also not allowed ("option score not
allowed")
Then I tried
est store first.
I did the same for the second xtnbreg equation.
Then I did
suest first second,
which yielded "score variables were not saved by model
new re-estimate with the score() option"
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://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/
