<>
Have you thought about the -fixedonly- option for -predict-?
**************
webuse epilepsy, clear
xtmepoisson seizures treat lbas lbas_trt lage visit || subject: visit
margins, predict(fixedonly) at(treat=(0 1) visit=(-.3(0.2).3))
**************
HTH
Martin
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of [email protected]
Sent: Montag, 21. September 2009 18:25
To: [email protected]
Subject: st: adjusted counts after xtmepoisson
I am running a crossed random effects model using xtmepoisson, after
which I want to generate estimated counts and confidence intervals at
different levels of X1 and X2. The documentation suggests that the
new "margins" command should be able to handle this -- it is listed as
an option after xtmepoisson -- but I get the error message: "default
prediction is a function of possibly stochastic quantities other than
e(b)". Using the "force" option only generates the error message:
"data have changed since estimation"
xtmepoisson y x1 x2 || _all:R_group_j || group_i:, mle
(output omitted)
margins, predict(mu) atmeans
Thanks!
*
* 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/
*
* 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/
