Re: [Gretl-users] Alternative way of formulating mle
On Fri, 27 Jun 2008, Gordon Hughes wrote:
Another query about mle. As before, I have been experimenting
with stochastic frontier models and I have found that mle
behaves entirely differently with two formulations that seem to
be equivalent...
I'm glad you spotted this; thanks for the report. This is now
fixed in CVS. (The issue was to do with the handling of NA values
generated by trying to take logs of non-positive values, in the
course of MLE iteration.)
Allin Cottrell