Your data analysis solution
Name is required.
Email address is required.
Invalid email address
Answer is required.
Exceeding max length of 5KB

Multinomial Logistic Regression Cannot be Estimated

By rxapur | May 11, 2016 10:22PM CEST

Hi,

I was attempting to run a multinomial logistic regression with 665 sample size and only 7 events. Got this resulting message "The maximum likelihood estimate may not exist. The multinomial logit model cannot be estimated."

Opted to use penalized likelihood (Firth's method) for alternative estimation to reduce the bias but it seemed that Firth's method is only available in binary logit model for xlstat.

I'd like to ask if there is any workaround or solution for this. Or maybe I'm doing something wrong or missed out on something. Thank you in advance.

Up

0

Down

By Efthalia | May 13, 2016 10:57AM CEST | XLSTAT Agent

Hello,

Please send your data and a brief description of your analysis to our support team via this form so that we can have a closer look at the problem.

Regards,

This question has received the maximum number of answers.

Contact Us

1c26995d494fb3061dd0ae8571ffc0a4@xlstat.desk-mail.com
https://cdn.desk.com/
false
desk
Loading
seconds ago
a minute ago
minutes ago
an hour ago
hours ago
a day ago
days ago
about
false
Invalid characters found
/customer/portal/articles/autocomplete
9283