Dear Olivia,
Thanks for your reply. I have been looking at the many files in the
./Zelig/R/ for some time and trying to figure out which file to begin
with; now, thanks to your replay, I know where to look.
Best,
Shige
On Wed, Nov 17, 2010 at 11:10 AM, Olivia Lau
<olivia.lau(a)post.harvard.edu> wrote:
Yes, there are two ways to do this.
(1) requires that you hack the output from the lme4a to be the same class as
the output from Zelig. Then run setx and sim as you normally would. This
is actually much more difficult than it sounds because lme4 and lme4a use S4
classes, which are not generally hackable.
(2) requires hacking the Zelig source code and recompiling the package.
This is easy with a Linux system, but much more difficult in Windows. If
you are up to the task, you start by downloading the Zelig_xxx.tar.gz (where
xxx is the current version number) from CRAN. Then go and find the
appropriate zelig2*** file in the R directory (where *** is the name of the
model). Where it says lme4 in that file, change it to lme4a. Compile, then
use Zelig as you normally would.
Good luck!
On Wed, Nov 17, 2010 at 8:20 AM, Shige Song <shigesong(a)gmail.com> wrote:
Dear All,
I am using zelig for estimating and simulating a mixed effect logit
model. The lme4 package, which Zelig used as the default, yield a
false convergence, while the experimental lme4a package yields
convergence and reasonable estimates. My question is: is it possible
to modify Zelig slightly to use lme4a instead of lme4 as the
underlying estimation engine for generalized linear mixed effect model
for this particular case? Many thanks.
Best,
Shige
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options:
http://lists.gking.harvard.edu/?info=zelig
Zelig program information:
http://gking.harvard.edu/zelig/
-
Zelig Mailing List, served by Harvard-MIT Data Center
Send messages: zelig(a)lists.gking.harvard.edu
[un]subscribe Options:
http://lists.gking.harvard.edu/?info=zelig
Zelig program information:
http://gking.harvard.edu/zelig/