Search found 1275 matches

by ChrisCharlton
Thu Aug 12, 2021 12:39 pm
Forum: runmlwin user forum
Topic: Bootstrap runmlwin
Replies: 21
Views: 1968

Re: Bootstrap runmlwin

If you wanted to avoid cross you could probably do the calculation in Mata, as in my previous example, but this is a bit more complicated. George's code already reduces the size of the dataset used in the prediction by throwing away duplicate combinations in the variables of interest (the duplicates...
by ChrisCharlton
Wed Aug 11, 2021 4:21 pm
Forum: runmlwin user forum
Topic: Bootstrap runmlwin
Replies: 21
Views: 1968

Re: Bootstrap runmlwin

Here are some answers to your questions from George: Why is thinning = 5? I understand that every fifth result is stored but I am not sure I understand why it is done. Should (burnin(#) and chain(#) be increased from the default? To reduce the size of the resulting saved file to make data manipulati...
by ChrisCharlton
Tue Aug 10, 2021 4:01 pm
Forum: runmlwin user forum
Topic: Bootstrap runmlwin
Replies: 21
Views: 1968

Re: Bootstrap runmlwin

George has had a look at this and provided us with an example for linear regression which you should be able to adapt (this is also simpler than mine and avoids using Mata): clear all * Open data use "http://www.bristol.ac.uk/cmm/media/runmlwin/tutorial", clear codebook schgend * Generate ...
by ChrisCharlton
Mon Aug 09, 2021 10:24 am
Forum: Realcom user forum
Topic: Imputation to level 2 variable: vary within group?
Replies: 5
Views: 333

Re: Imputation to level 2 variable: vary within group?

Thank you for confirming the cause of the error. I am curious as to where this apparent truncation is taking place. Would you be able to check the household id column in the file exported from Stata to see whether these high-valued IDs match the original data there? The file is loaded into Realcom-I...
by ChrisCharlton
Wed Aug 04, 2021 3:55 pm
Forum: runmlwin user forum
Topic: Bootstrap runmlwin
Replies: 21
Views: 1968

Re: Bootstrap runmlwin

One way to calculate a prediction with credible intervals, using your previous example would be as follows: clear all use http://www.bristol.ac.uk/cmm/media/runmlwin/bang, clear generate lc1 = (lc==1) generate lc2 = (lc==2) generate lc3plus = (lc>=3) foreach var of varlist hindu urban { bysort distr...
by ChrisCharlton
Wed Aug 04, 2021 2:26 pm
Forum: Realcom user forum
Topic: Imputation to level 2 variable: vary within group?
Replies: 5
Views: 333

Re: Imputation to level 2 variable: vary within group?

Realcom-Impute works out which responses are at each level by checking whether they vary within the provided level identifier. You could check that that the file generated by the Stata 'realcomimpute' command is correct by comparing it with the example on page 6 of the Realcom-Impute guide . The lev...
by ChrisCharlton
Wed Jul 28, 2021 12:53 pm
Forum: MLwiN user forum
Topic: Problem adding quadratic term
Replies: 2
Views: 396

Re: Problem adding quadratic term

I believe that this is probably due to saving and loading a worksheet containing a model in SPSS format. As the data in SPSS files has to be square MLwiN will fill in gaps with the "missing" value where this is not the case for the data being saved. You can see this from your screenshot wi...
by ChrisCharlton
Tue Jul 27, 2021 12:38 pm
Forum: MLwiN user forum
Topic: Multilevel Survival Analysis
Replies: 2
Views: 413

Re: Multilevel Survival Analysis

These macros have not been developed or tested for some time, but as far as I am aware these ought to still run on recent versions of the software with minimal modification. Here is a post from a recent attempt to use them a few years ago: viewtopic.php?t=2170
by ChrisCharlton
Tue Jul 27, 2021 11:50 am
Forum: runmlwin user forum
Topic: Bootstrap runmlwin
Replies: 21
Views: 1968

Re: Bootstrap runmlwin

I've asked George if he has any examples, but in the mean time it might be worth looking through the replication materials for the LEMMA course ( https://www.cmm.bris.ac.uk/lemma/ ), in particular modules 6 ( Regression models for binary responses ) and 7 ( Multilevel models for binary responses ) p...
by ChrisCharlton
Mon Jul 26, 2021 6:31 pm
Forum: runmlwin user forum
Topic: Bootstrap runmlwin
Replies: 21
Views: 1968

Re: Bootstrap runmlwin

To use the MCMC chains you would apply the normal prediction formula, except instead of single values for the parameter estimates you would use the chain of values instead. This would then give you a chain of predicted values where you can use the mean or median of the values as your prediction esti...