sfmatrix error

[Mike4kids]

I'm sure there must be ways to set the MLWin worksheet size from within Stata, but I can't figure it out.
Ran some toy examples, including ones provided. They worked, so stata is finding MLWin and so on.

I can see what is going on if I drop the nopause option, but then of course, I've kind of lost the benefit of running the program from within Stata.

What am I doing wrong?

[GeorgeLeckie]

Hi Mike,

runmlwin will automatically adjust the MLwiN worksheet size to cope with the specified model and dataset.

To override the default worksheet size (and some other settings) use the following option:

mlwinsettings(size())

See the runmlwin help file for full details

I hope this helps

George

[Mike4kids]

thanks for responding. I can see that I should have been able to figure out that setting on my own.
Of course, this begs the question of why I'm getting the error in the first place if MLWin adjusts the worksheet size automatically.

Making matters worse is that MLWin now crashes everytime when I launch it (whether through stata or directly
I have uninstalled and reinstalled several times --

[ChrisCharlton]

If MLwiN is now always crashing on startup then it is likely that you have set a default worksheet size that is too large. You can reset the size to its default by using the /reset command line option (see http://www.cmm.bris.ac.uk/forum/viewtopic.php?f=1&t=553).

[Mike4kids]

thanks, Chris--that fixed it --

now, any insights on the sfmatrix problem that started us down the path --

[ChrisCharlton]

If you can provide us with information on:

1) The exact error message that you are getting.

2) Details of the model that you are trying to run (possibly including the runmlwin syntax you are using).

3) Details about your data (number of cases, does it contain missing data, etc).

then we might be able to track down where it is going wrong.

[Mike4kids]

seems ok for now -- thank you both