Page 1 of 1

negative standard errors of random coefficients

Posted: 13 Nov 2022, 00:30
by duxdfx
I estimated a mixed logit model with three random coefficients (log-normal distributed). see below:
mu_log_b_prm -5.59949
sigma_log_b_prm -1.86121
mu_log_b_sub -9.10539
sigma_log_b_sub -4.87119
mu_log_b_lia -8.49144
sigma_log_b_lia 3.91300

Are these valid estimates?

Re: negative standard errors of random coefficients

Posted: 13 Nov 2022, 11:56
by stephanehess
Hi

this is entirely fine, please have a look at http://apollochoicemodelling.com/faq.html

Stephane