2 level model (logistic regression, IGLS)

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/
Post Reply
jelle
Posts: 3
Joined: Mon Apr 11, 2011 12:13 pm

2 level model (logistic regression, IGLS)

Post by jelle »

Hello

I'm trying to run a 2 level model with a binary response (logistic regression, IGLS). I have no problems when the predictors are level 1 variables (categorical or continuous). However, when trying to run models with level 2 predictors (all continuous data) I encounter the following problems.

Sometimes the model keeps running, so no B and SE are provided. A proposed solution was to allow the level 2 variance to become negative. This solved the problem in some cases. In other cases I recieved the following error messages:
'V gone -ve definite for block 2 iteration aborted (random pass) (moni)' OR '1.#QO'. In the forum I've read that the model might not work for all combinations of predictors / responses. Is this true? Is there no solution for this?

Furthermore, I wonder if it is OK to allow the variance to become negative as the obtained results differ from the results obtained in SPSS?

Thanks in advance!

Best regards

Jelle
Post Reply