www.cmm.bristol.ac.uk/forum

The variables after the iv are either the original predictors, or the corresponding instrument (and correspond to the IVSET column that you used in the MLwiN macro), so for you example you would want something like:

iv cons diffdist: runmlwin resp cons diffsch, level1(l1id: cons) ...
If you run ...

These shouldn't have a u error term unless you have added it by clicking on the variable and ticking the box for the second level, so to remove it you need to click on the again and remove the tick.

MLwiN does not have any special knowledge about what these macros are doing, so it won't do anything extra automatically for you.

FIXVARS and IVSET are new columns added to your data by the macros. The number in each row in FIXVARS corresponds to the column number of the corresponding variable in your model (for column numbers see the Data Manipulation->Names window). Initially the IVSET column is set to be the same as the ...

When you use the command PRINT 'FIXVARS' 'IVSET' it should just write the contents of these columns to the output window. To edit these you will instead have to do the following:

Data Manipulation->View or edit data
Click the "view" button
From the dropdown highlight FIXVARS and IVSET.
Click the ...

There is currently no way to specify alternative PRE and POST files when using runmlwin, so you won't be able to use these macros directly. I have however written a Stata wrapper that should do the equivalent of these particular macros (see http://www.cmm.bristol.ac.uk/forum/viewtopic.php?f=3&t=752 ...

If I understand the macro correctly the IVSET column that it creates has to have the same number of rows as there are fixed part parameters in the model. This message is telling you that this isn't the case for your model. When this happens the macro replaces the contents of IVSET with a row for ...

This column overwriting is the result of the bug in MLwiN v2.27 that I referenced here: http://www.cmm.bristol.ac.uk/forum/view ... =752#p1726. To use these macros you will either have to use a previous version (e.g. v2.26) or wait for the next update.

Would it be possible to provide a screenshot showing the issue, as I don't really understand the question?

Looking at chapter 19 it looks like the data does need to be in wide format, however it appears that multiple membership in IGLS also involves setting constraints. These would interfere with the constraints set by the ivmlwin macro, so unfortunately I don't think that you can use both of these at ...