Search found 156 matches

by billb
Thu Nov 28, 2013 9:34 am
Forum: MLwiN user forum
Topic: 3 level model: problems with cov matrix at level 2
Replies: 1
Views: 3150

Re: 3 level model: problems with cov matrix at level 2

Good Morning Claudia, It is as you have surmised possible for the IGLS method that really works on the full covariance matrix of the observations to construct parameter estimates that when evaluated as for example between school variances, correlations etc. give implausible values for this interpret...
by billb
Mon Nov 25, 2013 1:20 pm
Forum: MLwiN user forum
Topic: Multilevel modelling for categorical repeated measures data
Replies: 5
Views: 6933

Re: Multilevel modelling for categorical repeated measures d

Hi bgawarammana, To add to the last reply. Having spoken to colleagues this comment was made in quite an old version of the manual and has persisted. The main issue with repeated measures binary data is the assumption of normality for the individual random effects is often not satisfied by the data....
by billb
Mon Nov 25, 2013 11:41 am
Forum: MLwiN user forum
Topic: Multilevel modelling for categorical repeated measures data
Replies: 5
Views: 6933

Re: Multilevel modelling for categorical repeated measures d

Hi bgawarammana, It is possible to fit multilevel binary/categorical response models to data that have a structure of repeated measures within individuals. However just like all repeated measures models the assumption is that the correlation between individual data points is equal and doesn't depend...
by billb
Wed Aug 14, 2013 1:12 pm
Forum: MLwiN user forum
Topic: BIC with MLwin?
Replies: 1
Views: 12386

Re: BIC with MLwin?

Hi Sinavera, MLwiN only calculates the deviance (-2LL) though generally BIC is a simple formula involving this quantity and the number of parameters in the model and number of observations in the dataset. Things are not quite as simple in multilevel models as there is some disagreement onto what one...
by billb
Wed Jul 03, 2013 9:52 pm
Forum: MLwiN user forum
Topic: HELP setting up multiple membership model
Replies: 4
Views: 14857

Re: HELP setting up multiple membership model

Hi Diego, I think you are best transforming your data into long format and fitting a cross-classified model with exam score nested within pupil and school. That way you can easily include time varying covariates as you have 1 response per year. The alternative/extension where you make the school cla...
by billb
Tue Jul 02, 2013 9:35 am
Forum: MLwiN user forum
Topic: HELP setting up multiple membership model
Replies: 4
Views: 14857

Re: HELP setting up multiple membership model

Dear DrDTorres, My main concern here is whether your response is 1 per year ? If so then you simply have a simple 3 level model. If you want to do something more complicated then as I see it you have several choices. You could fit a variant of a cross-classified model which is in fact a 3 level nest...
by billb
Wed Jun 12, 2013 9:39 am
Forum: MLwiN user forum
Topic: multivariate Wald test
Replies: 1
Views: 8233

Re: multivariate Wald test

Dear mwkoeter, The intervals and tests window under Estimation will do this. You simply choose the fixed button at the bottom of the screen and increase # of functions to 3. Then for each column place a 1 next to 1 of the 3 dummies ( a different one for each column) and click calc to get the chi-squ...
by billb
Mon Mar 18, 2013 11:58 am
Forum: MLwiN user forum
Topic: spatial modeling
Replies: 1
Views: 4143

Re: spatial modeling

Dear user, I'll be honest with you and say that I am not sure I have ever tried a binary response model with spatial random effects. The majority of the time in disease mapping which is the field for which I coded the CAR modelling section the responses are treated as counts and fitted as Poisson. I...
by billb
Thu Jan 31, 2013 12:58 pm
Forum: MLwiN user forum
Topic: Multilevel Multiple Imputation
Replies: 2
Views: 5269

Re: Multilevel Multiple Imputation

Dear medha001, Just last week we came across a similar issue. The problem is if you have individuals who are missing on all responses then this causes issues with MLwiN when you switch to MCMC. We have made a change that will fix these issues, which will appear in the next version 2.27 but strictly ...
by billb
Fri Nov 16, 2012 8:24 am
Forum: MLwiN user forum
Topic: Changing values in Variance- Covariance matrix
Replies: 3
Views: 6679

Re: Changing values in Variance- Covariance matrix

Hi, In the situation in which you have level 1 variables that you make random at level 2 and for some reason IGLS gives a non-positive definite answer then the advice is to change the values prior to running MCMC so that they make sense. The values you choose have 2 roles - (i) starting values for t...