www.cmm.bristol.ac.uk/forum

If you are fitting the model using (R)IGLS then you could use a Wald test. If you are using MCMC then you could compare the values of the DIC diagnostic.

The most likely reason for this is that by default MLwiN applies an adjustment to the calculated variances to take into account sampling variability. This setting can be changed via the Model>Residuals window, or via the RTYPE command, however runmlwin does not currently have a way to set this. If ...

On the whole older macros should work with the most recent releases, so this may be a bug. I suggest updating to the most recent release of the software (available from https://www.cmm.bristol.ac.uk/clients/softwaredownload/ ), and if you still have the problem after this, providing further details ...

For cross-classified models the ordering of the residuals should match the sorted ID variable for the chosen classification. See the topic Residuals - matching for an MLwiN command to generate a new column that matches this.

Unfortunately we do not offer an automatic way to calculate these predictions. I think that your options are :


Remove the nopause option from the command and the once the model is estimated use Model>Customised predictions within MLwiN. You can then copy/paste these results back into Stata ...

There is some discussion of this in the MLPowSim manual. Multiple category predictors for MLwiN are covered in section 5.2, and for R in 5.4. You may in particular want to look at section 5.4.2 for an explanation.

The MISR command is used in conjunction with post-estimation calculation of the residuals and are not normally used in model estimation except with PQL1 or PQL2, or in MCMC where they are used for starting values. The runmlwin equivalent of this setting for post-estimation residuals is the norecode ...

If you run the model without the nopause option then MLwiN will pause at various points to allow you to inspect the model/data. You can can then resume it by clicking "Resume macro" on the toolbar. Once you have got through all the points that it pauses the results will be sent to Stata. When you ...

As this section of the forum is for questions related to the Stat-JR software package are you able to provide any more detail on how you are using it for this analysis, e.g. which templates and underlying statistics package you are using/planning to use?

You appear to have omitted your code, however if you want to specify CAR at more than one level you have to specify the carcentre option.