imputation model specification

Welcome to the forum for Stat-JR users. Feel free to post your question about Stat-JR here. The Centre for Multilevel Modelling take no responsibility for the accuracy of these posts, we are unable to monitor them closely. Do go ahead and post your question and thank you in advance if you find the time to post any answers!

We will add further support to the Stat-JR website, such as FAQs and tutorials, as soon as it is available; the Stat-JR website can be found here: http://www.bristol.ac.uk/cmm/software/statjr/
yongjookim78
Posts: 36
Joined: Tue Jan 10, 2017 3:36 am

Re: imputation model specification

Post by yongjookim78 » Thu Aug 17, 2017 2:22 pm

After completing the second point in the Step 3 (http://www.bristol.ac.uk/cmm/software/s ... er-statjr/),
I did the same procedure in Stat-JR as follows:

1. Template => "NLevelImpute"

2. Number of higher level classifications => 3
Higher level classification A: region (highest)
Higher level classification B: psun (second highest)
Higher level classification C: idfamn (second lowest)

3. Use conditional marginal algorithm? => Yes
3. Number of imputed data sets => 5
4. Number of iterations before first imputation => 1000
5. Number of iterations between subsequent imputations => 50
6. Length of burnin for MOI: => 100
7. Number of iterations for MOI: => 250

After I clicked "run", this time, it also worked shortly and the sign turned to "Ready(8s)" with nothing presented in the Stat-JR window.
The screen shot of the command window is attached! Would appreciate your help!!

Best wishes,
Yongjoo
Attachments
STATJR_screen1-1.png
STATJR_screen1-1.png (33.81 KiB) Viewed 37 times
STATJR_screen1-2.png
STATJR_screen1-2.png (69.28 KiB) Viewed 37 times
STATJR_screen1-3.png
STATJR_screen1-3.png (68.49 KiB) Viewed 37 times

yongjookim78
Posts: 36
Joined: Tue Jan 10, 2017 3:36 am

Re: imputation model specification

Post by yongjookim78 » Thu Aug 17, 2017 3:11 pm

I've got a couple more updates.

- I transited to a simpler template, "2LevelImpute" and realized that all categorical variables (ordered/unordered) should be coded from "0". Some of my variables were coded from "1". This was indicated when using "2LevelImpute" but not in "NLevelImpute."

- Then, based on "2LevelImpute", I did the same procedure after recoding all categorical variables to start from "0." It ran longer than before ("Ready(146s)" but still there is no change in the screen (i.e., no results showed up), as attached.

Would appreciate your helps!

Best,
Yongjoo
Attachments
STATJR_screen1-1.png
STATJR_screen1-1.png (58.39 KiB) Viewed 36 times
STATJR_screen1-2.png
STATJR_screen1-2.png (42.02 KiB) Viewed 36 times

ChrisCharlton
Posts: 870
Joined: Mon Oct 19, 2009 10:34 am

Re: imputation model specification

Post by ChrisCharlton » Wed Aug 23, 2017 9:21 am

The first of the two problems appears to be due to an error coming from one of the called templates. If you can provide a sample data set with the same structure that you are using I can try to look into this further, although it may be due to the category coding starting at one that you have identified.

The model in your second post looks like it has finished fitting. There should be a drop-down box containing output names just below the section that you provided a screenshot of. Is this missing, or are the no outputs there?

yongjookim78
Posts: 36
Joined: Tue Jan 10, 2017 3:36 am

Re: imputation model specification

Post by yongjookim78 » Thu Aug 31, 2017 1:28 pm

Thank you very much for all your helps Chris! Will try more and get it back to you again!
Best wishes,
Yongjoo

Post Reply