www.cmm.bristol.ac.uk/forum

The problem appears to be that there are three schools where all the observations contain some missing data. As these schools will be excluded from the model they are not counted in the hierarchy viewer.

You can find my email on the following web page:

http://www.bristol.ac.uk/cmm/team/

Would it be possible to email me with a copy of your worksheet (or at least the ID variables) so we can look into this further?

1) It is possible to disable the warning by clicking "Estimation control.." and then ticking "suppress numeric warnings". This only removes the messages however so your model will not converge any better than if you told it to continue estimating every time you get the message. 2...

You may want to try an undocumented experimental option in runmlwin to suppress the MLwiN GUI while running. To use this add the option batch where you would normally put nopause . The major downside to this is that you will no longer see any error messages produced by MLwiN, so it might be a good i...

There is no built-in function for doing this. Our recommended method for this is to carry out the analysis on standardised variables. This is not recommended for binary variables however (including dummies for categorical variables with more than two categories), as there is no meaningful interpreta...

I have now been able to replicate the error by reducing the size of my worksheet before running the example multinomial model, so it does indeed appear to be caused by the size of the data set. The incorrect error message should be fixed in the next release of MLwiN.

I have a feeling that the size of the data set may be what is causing the problem. As part of the MNOM command, where is it getting stuck, MLwiN expands the data to approximately number of response categories times the size. This may in your case be larger than the currently set MLwiN worksheet size...

Thanks for providing the macro. I cannot see anything obviously wrong with it. Would it be possible for you to email George or myself an example data set and do file that demonstrates the problem so that we can investigate further?

Would it be possible to add the viewmacro option to your runmlwin command and then post the contents of the generated script, as I would like to check that there is nothing obviously wrong there?