In case anyone is wondering, I figured out my problems.
1) I needed to set my dependent variable as a factor outside the zelig call.
2) It didn't like that I had missing data (which produced NAs in the setx values).
Listwise deletion prior to running the model solved that problem.
Cheers,
Evann
On Mar 29, 2013, at 6:25 PM, wrote:
Hi,
When running an ordered logit, I get the following error:
d1t.x <- setx(d1t)
d1t.s <- sim(d1t, x=d1t.x)
Error in as.matrix(x)[, -1] : incorrect number of dimensions
As best I can discern, this is because the coef(d1t) command pulls the coefficients from
the model, but not the values of the intercepts. While vcov(d1t) includes the intercepts.
Is there a fix or workaround for this? Or am I doing something else wrong?
Cheers,
Evann
====================================
Evann Smith
Ph.D. Candidate
Department of Government
Harvard University
egsmith@fas.harvard.edu<mailto:egsmith@fas.harvard.edu>
====================================
====================================
Evann Smith
Ph.D. Candidate
Department of Government
Harvard University
egsmith@fas.harvard.edu<mailto:egsmith@fas.harvard.edu>
====================================
Show replies by date