Search found 432 matches

by GeorgeLeckie
Mon Oct 24, 2011 11:54 am
Forum: runmlwin user forum
Topic: How do I stop MLwiN setting residuals to missing?
Replies: 4
Views: 6218

Re: How do I stop MLwiN setting residuals to missing?

Hi Corrie, Just to keep you informed, we have added an option ( norecode ) to runmlwin which will stop MLwiN from automatically recoding residuals which are predicted to be zero (or exceedingly close to zero) to be missing. We hope to release an updated version of runmlwin via SSC ( ssc install runm...
by GeorgeLeckie
Mon Oct 24, 2011 11:51 am
Forum: runmlwin user forum
Topic: Does runmlwin work with Stata 9 and Stata 10?
Replies: 3
Views: 5874

Re: Does runmlwin work with Stata 9 and Stata 10?

Dear Belkacem, Just to keep you informed, we have made some progress with making runmlwin work in Stata 9 and 10 as well as Stata 11 and Stata 12. We hope to release an updated version of runmlwin via SSC soon ( . ssc install runmlwin, replace ). We will let you and others know when we do so by post...
by GeorgeLeckie
Mon Oct 17, 2011 6:45 pm
Forum: runmlwin user forum
Topic: How do I stop MLwiN setting residuals to missing?
Replies: 4
Views: 6218

Re: How do I stop MLwiN setting residuals to missing?

Hi Corrie,

As a temporary workaround, you could simply manually recode residuals with missing values on their return to Stata?

Code: Select all

. recode u0 u1 (.=0) if e(sample)==1
Best wishes

George
by GeorgeLeckie
Fri Oct 14, 2011 1:58 pm
Forum: runmlwin user forum
Topic: MCMC Error 0315: Prior variance matrix is not positive defin
Replies: 3
Views: 8086

Re: MCMC Error 0315: Prior variance matrix is not positive d

Dear Raphael, Thank you very much for your useful post. I am sure other users will benefit from this. In terms of initial values, as long as they are reasonably credible and as long as you run for a suffficiently long burnin then you should be fine. So in you case specifying variances of 0.1, a cova...
by GeorgeLeckie
Thu Oct 13, 2011 12:08 pm
Forum: runmlwin user forum
Topic: MCMC Error 0315: Prior variance matrix is not positive defin
Replies: 3
Views: 8086

Re: MCMC Error 0315: Prior variance matrix is not positive d

Hi Rapheal, When MLwiN says "MCMC Error 0315: Prior variance matrix is not positive definite" what this is telling you is that the IGLS estimates for one or more of your variance covariacne matrices are dodgy. Typically the problem is that the correlations implied by your covariance matrix...
by GeorgeLeckie
Thu Oct 13, 2011 11:51 am
Forum: runmlwin user forum
Topic: 3-level multinomial error:'illegal op: v_block:: mult_add()'
Replies: 17
Views: 20735

Re: 3-level multinomial error:'illegal op: v_block:: mult_ad

You're very welcome,

Glad runmlwin is now working for you.

George
by GeorgeLeckie
Thu Oct 13, 2011 11:46 am
Forum: runmlwin user forum
Topic: Does runmlwin work with Stata 9 and Stata 10?
Replies: 3
Views: 5874

Re: Does runmlwin work with Stata 9 and Stata 10?

Dear Belkacem,

Unfortunately runmlwin currently only works with Stata 11 and Stata 12.

We will look into what is required to make it backwards compatible with Stata 10 and earlier versions and then post what we intend to do about backwards compatibility here shortly.

Best wishes

George
by GeorgeLeckie
Tue Oct 11, 2011 7:27 pm
Forum: runmlwin user forum
Topic: How do you constrain covariances to zero?
Replies: 2
Views: 4911

Re: How do you constrain covariances to zero?

Hi Corrie, Yes there was a small error in the constraints option. Thank you for spotting this. We have now fixed this and this will be available officially in a week or so. We have emailed you a temporary file to fix the problem while you wait for the official released. However, we were not able to ...
by GeorgeLeckie
Mon Oct 10, 2011 6:33 pm
Forum: runmlwin user forum
Topic: 3-level multinomial error:'illegal op: v_block:: mult_add()'
Replies: 17
Views: 20735

Re: 3-level multinomial error:'illegal op: v_block:: mult_ad

You have ~700,000 individuals, and you want to fit a three-level three-category multinomial response model. The expanded data is therefore ~1,400,000. My gut feeling is that this is too large for MLwiN (or other packages) to handle and that you should work with a random sub-sample of your data. Many...
by GeorgeLeckie
Fri Oct 07, 2011 9:28 am
Forum: runmlwin user forum
Topic: 3-level multinomial error:'illegal op: v_block:: mult_add()'
Replies: 17
Views: 20735

Re: 3-level multinomial error:'illegal op: v_block:: mult_ad

Fundamentally you have a very large data set (700,000 individuals) and an unordered categorical response variable. MLwiN and other statistical packages are always going to struggle to fit a three-level model to these data. I think you will have to work with a smaller sub sample of these data, at lea...