Search found 1254 matches

by ChrisCharlton
Fri Jul 23, 2021 12:36 pm
Forum: runmlwin user forum
Topic: Bootstrap runmlwin
Replies: 7
Views: 179

Re: Bootstrap runmlwin

If you could adapt your example to a dataset that I could use to replicate this (for example the MLwiN tutorial data) then I can attempt to work out where this error is coming from.
by ChrisCharlton
Fri Jul 16, 2021 2:24 pm
Forum: Realcom user forum
Topic: Help me please
Replies: 1
Views: 87

Re: Help me please

Unfortunately there is not an easy way to set the starting values when running models in Realcom. I have not tested this but it appears that if you place a file named L2VARSTART into the Realcom-Impute directory then the values in this will be loaded as the starting values for the level-2 variances,...
by ChrisCharlton
Wed Jul 14, 2021 3:53 pm
Forum: runmlwin user forum
Topic: Keep data in MLwiN after model estimation to speed up subsequent modeling?
Replies: 2
Views: 113

Re: Keep data in MLwiN after model estimation to speed up subsequent modeling?

Unfortunately this functionality it not available, as the command requires MLwiN to be in a predictable state before it fits each model. You can however use the viewfullmacro , savemacro or saveworksheet options of the command to extract the required syntax, or the worksheet used to fit the model in...
by ChrisCharlton
Mon Jul 12, 2021 10:24 am
Forum: runmlwin user forum
Topic: Bootstrap runmlwin
Replies: 7
Views: 179

Re: Bootstrap runmlwin

This is not something that we are particularly familiar with, however the examples I found seem to add rclass rather than eclass to the bootstrap program and appear to require you to pass on any required outputs from the estimation command using return . Bootstrap sampling and estimation How do I wr...
by ChrisCharlton
Fri Jul 09, 2021 1:37 pm
Forum: runmlwin user forum
Topic: Bootstrap runmlwin
Replies: 7
Views: 179

Re: Bootstrap runmlwin

I checked this with George, and he recommended that you would probably want to change the runmlwin estimation option to be pql2 instead of the default mql1 for this.
by ChrisCharlton
Fri Jul 09, 2021 12:49 pm
Forum: runmlwin user forum
Topic: Help!!!
Replies: 1
Views: 81

Re: Help!!!

The recommendation for comparing these types of models is usually to use a Wald test, or if you are estimating via MCMC to compare the DIC statistics.
by ChrisCharlton
Fri Jul 09, 2021 12:46 pm
Forum: MLwiN user forum
Topic: A small query
Replies: 1
Views: 70

Re: A small query

You should be able to set the number of levels in the drop-down box at the top of the dialogue box where you set the response and level identifiers (you can get to this by clicking on your response in the equation). If this doesn't offer the number of levels that you require then you can increase th...
by ChrisCharlton
Mon Jul 05, 2021 2:09 pm
Forum: MLwiN user forum
Topic: Issue: 1 out of 508 in use (lev 2 var) - LEMMA Mod 5
Replies: 1
Views: 49

Re: Issue: 1 out of 508 in use (lev 2 var) - LEMMA Mod 5

This sometimes happens when general notation is turned off and the cons variable, used to represent the intercept is added to the model before any other variables. As it doesn't yet know the length of the data being used in the model this is instead created with a length 1 and will subsequently resu...
by ChrisCharlton
Wed May 19, 2021 2:00 pm
Forum: runmlwin user forum
Topic: Area Under Curve (AUC) after random Effects models
Replies: 1
Views: 5172

Re: Area Under Curve (AUC) after random Effects models

I checked this with George, and his advice was to firstly calculate the predicted probability in the usual way and then to use the roctab command to obtain the AUC from this.
by ChrisCharlton
Wed May 19, 2021 11:47 am
Forum: R2MLwiN user forum
Topic: Issues when storing residual chains
Replies: 2
Views: 7407

Re: Issues when storing residual chains

Would it be possible for you to confirm that the following, which should be an equivalent model using the MLwiN example data works for you? > library(R2MLwiN) > data(bang, package = "R2MLwiN") > f <- logit(use) ~ 1 + age + (1 | district) > m <- runMLwiN(f, D = "Binomial", data = ...