if you change the starting values or tighten up on the priors a bit you
can usually get it to converge in difficult cases. I'd try that.
Gary
: Gary King, King(a)Harvard.Edu
http://GKing.Harvard.Edu :
: Center for Basic Research Direct (617) 495-2027 :
: in the Social Sciences Assistant (617) 495-9271 :
: 34 Kirkland Street, Rm. 2 HU-MIT DC (617) 495-4734 :
: Harvard U, Cambridge, MA 02138 eFax (928) 832-7022 :
On Thu, 23 Oct 2003, Gregory Pettis wrote:
For my dissertation I've estimated nine seperate
EI model specifications
for 9 seperate U.S. states, with most of the models employing a variety of
covariates.
All of the models' ML functions converged normally, except for two of the
states. In the first state 3 of the 9 converged, and in the second 5 of
the nine. One of my advisors has instructed me to explain why in these
two states the models failed to converge. The failure to converge does
not seem to be dependent on the numbers of covariates. Obviously the
models without covariates converged, as did the models with the most
numbers of covariates. In one of the states none of the models with the
covariates which were dichotomized converged, and in the other state
convergence was not dependent on how the covariates were coded.
Also, since there was no convergence, I don't have any output to evaluate.
What do you suggest?
Greg
Gregory A. Pettis
Political Science
Elon University
CB # 2203
Elon, N.C. 27244
-
EI mailing list served by Harvard-MIT Data Center
Subscribe/Unsubscribe:
http://lists.hmdc.harvard.edu/?info=ei
-
EI mailing list served by Harvard-MIT Data Center
Subscribe/Unsubscribe:
http://lists.hmdc.harvard.edu/?info=ei