I'm trying to work my way through multiple imputation. If I don't define constraints, I get some negative numbers which don't make sense (as my data can't be negative!). However, If I do define constraints by setting the minimum values to zero, I'm getting a message that reads "After 100 draws, the imputation algorithm cannot find an imputed value under the constraints available for variable ...... Check the minimum and maximum values specified to determine that they re reasonable, and consider raising the number of draws allowed". I have raised the number of draws to 500, 1000, 2000, but it still does it.
Please suggest how to proceed.
Thanks.
You should consider posting your question to the SPSSX-L mailing list. The easiest way to access it, I think, is via the Nabble page. The second link below is for a page that allows you to join the mailing list. If you don't like e-mail lists, you can set the nomail option when you join, and interact with the list via the Nabble website. HTH.
http://spssx-discussion.1045642.n5.nabble.com/
http://spssx-discussion.1045642.n5.nabble.com/mailing_list/MailingListOptions.jtp?forum=1068821
You might want to run a custom model using Predictive Mean Matching (PMM) instead of linear regression (negative numbers after MI might be due to skewed data).
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS