Search found 1242 matches

by ChrisCharlton
Fri Dec 04, 2020 12:23 pm
Forum: runmlwin user forum
Topic: Three-level model with two random slopes
Replies: 1
Views: 798

Re: Three-level model with two random slopes

I asked George about this and he said the following: It is harder to fit random slopes in models with a binary response, and it is easier to hit convergence issues. There might not be sufficiently large clusters and response variation to fit the desired model. In this case it appears that the level-...
by ChrisCharlton
Thu Dec 03, 2020 1:13 pm
Forum: runmlwin user forum
Topic: Order of Predicted Residuals with Multiple Random Slopes
Replies: 2
Views: 795

Re: Order of Predicted Residuals with Multiple Random Slopes

These should be the order that the parameters appear in the model equation. You can see this by removing the nopause option so that this is displayed prior to fitting the model. This ordering will usually match the order in which runmlwin processes the variables, so in your case as you do not have c...
by ChrisCharlton
Wed Dec 02, 2020 2:14 pm
Forum: Stat-JR user forum
Topic: Error running 2LevelUnordered Template in MCMCglmm and Rrstan
Replies: 6
Views: 1716

Re: Error running 2LevelUnordered Template in MCMCglmm and Rrstan

Adding the as.factor() call into R should be fine as an alternative to having the variable converted when loaded into R. You'd just have to make sure that the variable is suitable for this yourself (i.e. contains a set of discrete integer values). The advantage of using value labels is that the diff...
by ChrisCharlton
Mon Nov 30, 2020 4:05 pm
Forum: runmlwin user forum
Topic: Increase transmission speed from runmlwin to Stata
Replies: 4
Views: 1145

Re: Increase transmission speed from runmlwin to Stata

If you look at the runmlwin help via the following Stata command: help runmlwin You will see that there are a number of options for controlling the display under the Reporting options. You may find that you can speed up the display by changing some of these (for example omitting the fixed-effects ta...
by ChrisCharlton
Mon Nov 30, 2020 12:18 pm
Forum: runmlwin user forum
Topic: Increase transmission speed from runmlwin to Stata
Replies: 4
Views: 1145

Re: Increase transmission speed from runmlwin to Stata

You could try using the mlnscript.exe file instead of mlwin.exe when calling MLwiN, as this will avoid setting up the GUI interface each time you call the software. The downside is that you won't be able to see the model equation, etc or fitting progress as the model runs.
by ChrisCharlton
Fri Nov 27, 2020 1:20 pm
Forum: Stat-JR user forum
Topic: Error running 2LevelUnordered Template in MCMCglmm and Rrstan
Replies: 6
Views: 1716

Re: Error running 2LevelUnordered Template in MCMCglmm and Rrstan

Looking at the template history it looks like we removed a question specifying a denominator variable as none of the packages called by the template support this, however we appear to have missed removing a piece of code that assumes that this has been specified. I believe that if you remove the fol...
by ChrisCharlton
Mon Nov 23, 2020 2:59 pm
Forum: runmlwin user forum
Topic: mlwin on a cluster
Replies: 1
Views: 721

Re: mlwin on a cluster

In general MLwiN is a serial process, however it does use the BLAS libraries so it you have large enough data and link to a parallel version of these you may gain some benefit. If you have several models to fit, or want to run multiple MCMC chains then you could take advantage of the extra processor...
by ChrisCharlton
Mon Nov 23, 2020 2:49 pm
Forum: Stat-JR user forum
Topic: Error running 2LevelUnordered Template in MCMCglmm and Rrstan
Replies: 6
Views: 1716

Re: Error running 2LevelUnordered Template in MCMCglmm and Rrstan

For your first model it looks like there was a problem generating the R script: Error: unexpected symbol in "Traceback (most recent" Execution halted Would you be able to post what is generated (it should be in an output with a name like script.R)? For the Stan model it looks like the data is not be...
by ChrisCharlton
Wed Nov 11, 2020 5:01 pm
Forum: MLwiN user forum
Topic: Issue running MLPowSim in R
Replies: 2
Views: 672

Re: Issue running MLPowSim in R

What happens if you just remove the method="ML" statement from the command? i.e.:

Code: Select all

lmer(modelformula, data)
If that doesn't make a difference could you post the associated generated R code?
by ChrisCharlton
Wed Nov 11, 2020 4:47 pm
Forum: runmlwin user forum
Topic: Multilevel multinomial logistic model with survey weights
Replies: 2
Views: 868

Re: Multilevel multinomial logistic model with survey weights

Unfortunately, as indicated by the message, you can only specify weights for models that are both fitted with (R)IGLS and have a univariate response via runmlwin.