multinomial logistic regression - Error design vector length - no missings

Welcome to the forum for MLwiN users. Feel free to post your question about MLwiN software here. The Centre for Multilevel Modelling take no responsibility for the accuracy of these posts, we are unable to monitor them closely. Do go ahead and post your question and thank you in advance if you find the time to post any answers!

Remember to check out our extensive software FAQs which may answer your question: http://www.bristol.ac.uk/cmm/software/s ... port-faqs/
ChrisCharlton
Posts: 1351
Joined: Mon Oct 19, 2009 10:34 am

Re: multinomial logistic regression - Error design vector length - no missings

Post by ChrisCharlton »

It is hard to determine what is going on here, especially without seeing the data, however it may be related to around half the data being in one of the categories (NRM Only). Because of this there may be groups where all of the records are in this category. If MQL1 works and PQL2 doesn't then I would suggest looking at the level-2 residuals for the MQL1 model as the residuals are used for the PQL2 estimation.

We often suggest running these type of model using MCMC, have you tried this and if so do the results look sensible?
Tunze123
Posts: 3
Joined: Mon Oct 28, 2019 5:22 am

Re: multinomial logistic regression - Error design vector length - no missings

Post by Tunze123 »

Thanks Charlton, I formed a binary of the variable and it worked. Thanks for your assistance
Post Reply