Search found 156 matches

by billb
Tue Sep 20, 2016 11:53 am
Forum: MLwiN user forum
Topic: effect sizes?
Replies: 3
Views: 9101

Re: effect sizes?

Effect sizes come in many shapes and sizes and are often some function of the parameters of a statistical model or even simply the parameters themselves - for example Cohen's d is a measure based on the parameters for a difference scaled by standard errors and is very similar to a standardised regre...
by billb
Fri May 27, 2016 1:35 pm
Forum: MLwiN user forum
Topic: How to use first observation in growth model simulation
Replies: 3
Views: 4917

Re: How to use first observation in growth model simulation

Hi kaiserdominici, So it's clear that what you tried shouldn't work i.e. you removed years 2 and 3 for everyone so there is no data to estimate anything about them. In theory if you had done this for only a few individuals then the model should at least work but MCMC in MLwiN is only designed to &qu...
by billb
Tue May 03, 2016 1:45 pm
Forum: MLwiN user forum
Topic: multilevel-analysis, predictor
Replies: 4
Views: 5439

Re: multilevel-analysis, predictor

Hi Ine, One always has to model at the level of the response i.e. if you are looking at what factors are affecting something measured at the round level then the predictors have to also be at the round level. If you have predictors that are measured at a finer granularity then you have to aggregate ...
by billb
Mon Apr 04, 2016 2:55 pm
Forum: MLwiN user forum
Topic: MCMC, (customised) predictions & their interpretation
Replies: 1
Views: 3331

Re: MCMC, (customised) predictions & their interpretation

Hi Harmen, Can you clarify your data structure for me as you refer to both cross-classification and multiple memberships! You mention people and countries so my question is at what level are your observations taken and then from that information I may be able to help further. Best wishes, Bill.
by billb
Wed Mar 09, 2016 9:10 am
Forum: runmlwin user forum
Topic: 4 level cross-classified model estimation
Replies: 3
Views: 4996

Re: 4 level cross-classified model estimation

Hi Viacheslav, Thanks for your email. So to clarify nesting is indeed a special case of the more general relationship that we called crossed-classification or perhaps better called multiple classifications where a response depends on a series of classifying variables. The general case is you have se...
by billb
Mon Feb 15, 2016 11:16 am
Forum: MLwiN user forum
Topic: Multiple membership multiple classification (MMMC) or Cross classification (CC) model ?
Replies: 1
Views: 3368

Re: Multiple membership multiple classification (MMMC) or Cross classification (CC) model ?

Hi Zhaohaiyan,
Yes given you have a different response nested within a rater and a student it seems to me that XC is more appropriate. MMMC is generally for 1 response shared between the set of raters.
Hope that helps,
Bill.
by billb
Thu Aug 06, 2015 10:50 am
Forum: MLwiN user forum
Topic: Binary MLM-Level 1 variance
Replies: 1
Views: 10421

Re: Binary MLM-Level 1 variance

Hi Stasis, Not all single level logistic regression techniques translate directly to the multilevel case. Even in normal models the concept of R^2 is not usually referred to in multilevel situations as you do not have a single variance to reduce! The multilevel logistic also has interesting properti...
by billb
Thu Aug 06, 2015 10:41 am
Forum: MLwiN user forum
Topic: Binary MLM assumptions
Replies: 1
Views: 10517

Re: Binary MLM assumptions

Hi Stasis, At higher levels in the model you are assuming normality of residuals so you can test normality here as per normal models. Not much point at looking at level 1 residuals here as they often form 2 bands - one for the 1s and one for the 0s! I guess the concept of outliers is difficult when ...
by billb
Thu Jul 02, 2015 4:02 pm
Forum: MLwiN user forum
Topic: MCMC variance overestimation
Replies: 3
Views: 17595

Re: MCMC variance overestimation

Dear Anne-Marie,
If you were able to send me the worksheet I'd happily take a look and check that there wasn't anything silly going on.
Best wishes,
Bill.
by billb
Wed May 27, 2015 7:23 am
Forum: MLwiN user forum
Topic: VPC ordered multinomial (logit)
Replies: 1
Views: 16413

Re: VPC ordered multinomial (logit)

Hi Petra, MLwiN fits multinomial models in a somewhat different way to a standard binomial logistic model - in particular it uses an additional level below individual to capture the pattern of the multinomial response (in the ordered case a series of 1s and 0s indicating which threshold the true val...