Page 1 of 1

Conditional Logit

Posted: Wed May 21, 2014 7:23 am
by Jerak
Hi all,

I have to estimate the parameters of a generalized conditional logit model by maximum likelihood. In the model both dummy and continious variables are present. I have identified the parameters, however the standard errors and probabilities of the coefficients are 'NA'. I get the error: WARNING: Singular covariance - coefficients are not unique. I have identified several combinations of parameters, but it stays the same.

Could anyone help me solving this problem?

Thanks in advance.

Attached, the workfile.