www.cmm.bristol.ac.uk/forum

I'm trying to run some 2- and 3-level binary models via runmlwin. For some reason, MLwiN is taking a very long time to load the data each time a model is run — the "Reading record..." message in the status bar takes about 10 minutes to finish. I've run similar models without problems in the past ...

Great – thanks for your help with this, it's much appreciated.

I've done as you suggested:


Estimate the model in MLwiN (via runmlwin)
Set all variables to 0, except those we need for prediction.
Generate predicted values and standard errors (using predict )
Plot the graph, as before.

This ...

Update: the above syntax worked perfectly.. For example:

https://img.skitch.com/20110727-bqbp15d8qppm8d15ewksrr5g99.png

For those interested, the syntax to achieve this is:

xi: runmlwin satlife cons ///
incgm agegm female i.genhlth2 socact socmeet ///
discuss i.feelinc2 ptnr children i ...

Perfect! I'll give that a try tomorrow and let you know how it goes..

Ewan
--

Apologies if this is a stupid question: my lack of Stata/MLwiN knowledge isn't helping here...

Question: How can I get predictions from MLwiN via runmlwin ?

For example, using the predictions window I can select several variables, choose a column,
and then output the predicted values to that ...

Thanks for the quick reply. I'll try the compress command.

Ewan
--

First of all: THANK YOU! runmlwin will literally save me hours and
hours (not to mention the likely RSI caused by repeatedly clicking
through MLwiN's interfaces) :)

I have a question: is there something I can do to prevent MLwiN giving me
the following message each time a runmlwin command is run ...

Hi all,

I'm estimating a two-level logistic regression model, with MCMC estimation. Previously, I've used IGLS estimation, and have used "Model >> Residuals" to plot the ranked residuals for each level-2 unit (the typical 'caterpillar plot').

I notice in the MCMC manual (Browne, 2009) that there ...