Search found 112 matches

by billb
Mon May 21, 2018 10:28 am
Forum: runmlwin user forum
Topic: Can runmlwin handle CAR models at two levels?
Replies: 3
Views: 1500

Re: Can runmlwin handle CAR models at two levels?

Hi Yransome, The CAR modelling in MLwiN (and hence runMLwiN) is limited to only having a constant random at the CAR distributed residuals level i.e. we can't include a spatially varying random slope in a CAR model. We did extend the CAR modelling so that you could have different levels of CAR residu...
by billb
Sat May 12, 2018 7:26 am
Forum: MLwiN user forum
Topic: MLPowSim macro not working in MLWin
Replies: 3
Views: 1371

Re: MLPowSim macro not working in MLWin

Hi Adam80090,
I'll pass that on to my colleague Chris Charlton as I am sure we can probably fix this in the software - I wrote the MCMC functionality but not the core code so this one will be in Chris's domain.
Thanks for pointing it out.
Bill.
by billb
Tue Apr 24, 2018 10:48 am
Forum: MLwiN user forum
Topic: Heteroscedasticity and robust standard errors
Replies: 1
Views: 1146

Re: Heteroscedasticity and robust standard errors

Hi,
I am not sure precisely what model and what robust ses you refer to here. Generally in MCMC one would look at credible intervals if one wanted to get confidence intervals that are not restricted to normality so perhaps just use them.
Bill.
by billb
Mon Apr 16, 2018 8:56 am
Forum: MLwiN user forum
Topic: Testing Nested Models
Replies: 2
Views: 1333

Re: Testing Nested Models

Hi Nick, That's quite tricky. If you are using classical estimation then you can do an approximate Wald test though this isn't so great as assumes normality for the variance parameter. Alternatively you can use the DIC diagnostic with MCMC estimation although again this is not universally liked but ...
by billb
Thu Apr 12, 2018 2:33 pm
Forum: MLwiN user forum
Topic: Confidence intervals for variance function
Replies: 3
Views: 1586

Re: Confidence intervals for variance function

Hi Jerome, If you look at my MCMC book then in section 4.9 I describe how one works out estimates and confidence intervals for functions of parameters - in this case a VPC and so you could do something similar. Perhaps an even easier solution is to to reparameterise your model so that you don't have...
by billb
Thu Apr 12, 2018 1:44 pm
Forum: MLwiN user forum
Topic: Confidence intervals for variance function
Replies: 3
Views: 1586

Re: Confidence intervals for variance function

Hi Jerome, It's not usually the case that variances follow normal distributions but you would get an approximate interval using the SE. If you used MCMC estimation then you could get a chain of variance functions and thus an empirical credible interval for the 2 variances. Don't know if that is help...
by billb
Thu Feb 22, 2018 8:42 am
Forum: MLwiN user forum
Topic: MLPowSim macro not working in MLWin
Replies: 3
Views: 1371

Re: MLPowSim macro not working in MLWin

Hi Michael, Thanks for the query. Always good to hear people using MLPowSim even if they aren't having much success. I can't really diagnose the issue as you have only sent some of the files through and I suspect setup.txt is the one to look at. I wonder whether the issue is the number of state cate...
by billb
Wed Jan 24, 2018 2:08 pm
Forum: R2MLwiN user forum
Topic: Relatively simple MCMC logistic model needs many iterations
Replies: 3
Views: 1391

Re: Relatively simple MCMC logistic model needs many iterations

My email address is on the CMM website team page.
Best wishes,
Bill.
by billb
Wed Jan 24, 2018 2:06 pm
Forum: R2MLwiN user forum
Topic: How to check hierarchy in complex cross-clasified structure
Replies: 3
Views: 1518

Re: How to check hierarchy in complex cross-clasified structure

Hi Gesine, The point is that they don't need to be recognized for the model to fit correctly - again you should have unique IDs so don't have classT1 = 1 in school 1 and classT1 = 1 in school 2 or they'll be treated as the same class. The MCMC algorithm unlike the IGLS algorithm doesn't rely on nest...
by billb
Wed Jan 24, 2018 1:47 pm
Forum: R2MLwiN user forum
Topic: How to check hierarchy in complex cross-clasified structure
Replies: 3
Views: 1518

Re: How to check hierarchy in complex cross-clasified structure

Hi Gerine, If you are using MCMC for your cross-classified model then the software doesn't need to know nestings as nested classifications are just a special case of the more general crossed classification. The only thing you have to be careful of is with regard how you label your units so things wi...