Search found 1335 matches

by ChrisCharlton
Mon Oct 10, 2022 1:41 pm
Forum: runmlwin user forum
Topic: Endogeneity issues
Replies: 3
Views: 12601

Re: Endogeneity issues

That would be my interpretation of George's suggest as well.
by ChrisCharlton
Mon Oct 10, 2022 1:39 pm
Forum: R2MLwiN user forum
Topic: bug when setting starting parameters because of R update
Replies: 3
Views: 21284

Re: bug when setting starting parameters because of R update

Thank you very much for reporting this problem. It should be fixed in the development version of the package, available from https://github.com/r-forge/r2mlwin/tree/master/R2MLwiN . We are hoping to do a new CRAN release that contains this and some other fixes soon, however we still have to do a few...
by ChrisCharlton
Tue Aug 16, 2022 9:11 am
Forum: runmlwin user forum
Topic: Endogeneity issues
Replies: 3
Views: 12601

Re: Endogeneity issues

I asked George about this and his reply was as follows: I guess you could apply ideas from fixed versus random effects to panel data which are motivated by endogeneity concerns. So you could enter region as fixed-effects dummy variables instead of as random effects. The effects of the firm covariate...
by ChrisCharlton
Tue Jul 19, 2022 12:31 pm
Forum: runmlwin user forum
Topic: svyset compatibility with runmlwin
Replies: 1
Views: 4483

Re: svyset compatibility with runmlwin

I asked George about this and he suggested that the weighting options in MLwiN might do what you want. Unfortunately we don't provide much documentation for this, but you can see how to specify them by looking at the options in the runmlwin help file, i.e.

Code: Select all

help runmlwin
by ChrisCharlton
Mon Jul 04, 2022 2:36 pm
Forum: MLwiN user forum
Topic: 2 responses using MLwiN3.01
Replies: 1
Views: 3779

Re: 2 responses using MLwiN3.01

The reason for this is that your have chosen Binomial responses. This response type has a fixed level-1 variance (level-2 in this case as everything is shifted up a level for multivariate responses), which therefore cannot be estimated. If you had fitted a univariate Binomial response you would find...
by ChrisCharlton
Fri Jun 17, 2022 1:51 pm
Forum: MLwiN user forum
Topic: Calculating level-three residuals in multilevel model
Replies: 2
Views: 3940

Re: Calculating level-three residuals in multilevel model

If you are using MLwiN then the residuals returned from the Model>Residuals menu are already shrunken (see chapter 3 of the MLwiN User's Guide ). If you need an unshrunken version then the REFLATE command (see the MLwiN on-line help) is provided to calculate this, and more details of the formula use...
by ChrisCharlton
Tue Jun 14, 2022 11:05 am
Forum: R2MLwiN user forum
Topic: informative priorParam in random slope
Replies: 2
Views: 3575

Re: informative priorParam in random slope

As you state the 'prior variance matrix is not positive definite' message usually occurs when the starting values for the covariance matrix, used as the prior for MCMC, is invalid. Normally this matrix is obtained from the IGLS estimates, however you can specify it manually via the startval option. ...
by ChrisCharlton
Tue Apr 12, 2022 1:14 pm
Forum: runmlwin user forum
Topic: Level 2 unit problem
Replies: 1
Views: 4633

Re: Level 2 unit problem

Although you have checked the sorting, this does look like the sort of behaviour you would get if the data was not sorted by the identifiers defining the unit hierarchy. Can you confirm that the data is sorted by all of these ('n', 'wave_num' and 'wave')? You could also try recoding 'n' to go from 1...
by ChrisCharlton
Thu Mar 10, 2022 9:04 pm
Forum: R2MLwiN user forum
Topic: how to extract credible intervals
Replies: 1
Views: 3719

Re: how to extract credible intervals

For MCMC models the parameter chains are stored in a chains slot of the returned object, in your case this would be modelstim@chains . You can use the functions provided in the coda package to manipulate these, for example: library(coda) summary(modelstim@chains) will list summary statistics for eac...
by ChrisCharlton
Fri Feb 25, 2022 1:41 pm
Forum: runmlwin user forum
Topic: melogit verse runmlwin
Replies: 4
Views: 5254

Re: melogit verse runmlwin

I asked George if he could think of an explanation for this and he suggested running the model with PQL2 instead of MQL1, as MQL1 variances are biased downwards.