Announcement

Collapse
No announcement yet.
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Multivariate imputation

    Thank you very much for clicking in.
    I am trying to conduct multivariate imputation using commands: mi impute chained and mi impute mvn. But neither works well. Below is the outcome I got:

    Click image for larger version

Name:	1.png
Views:	1
Size:	85.9 KB
ID:	1654754

    Click image for larger version

Name:	2.png
Views:	1
Size:	285.2 KB
ID:	1654755
    I
    Would you please tell me where went wrong?
    Many thanks for your help.

  • #2
    Have you checked whether all auxiliary variables have no missing data?

    Comment


    • #3
      Originally posted by Tiago Pereira View Post
      Have you checked whether all auxiliary variables have no missing data?
      Thank you! Yes, I have checked all the other variables. And use mi impute regular to mark those variables. Actually, I have tried:
      mi impute chained (poisson) date place recalli ser74= r4agey gender marriage h4child livealone edu1 edu2 hukou hten income chronic ADLsum IADLsum needsum [pweights=r4wtresp], add(10)
      still not working.
      Date, place, recall and ser74 are four variables with missing data, and the rest of them are employed in analysis with no missing data.

      Comment


      • #4
        I am very confused about the weight under the syntax. The result showed that weights were not allowed. But the help mi impute document showed the opposite.
        Click image for larger version

Name:	1.png
Views:	1
Size:	137.1 KB
ID:	1654770
        Click image for larger version

Name:	2.png
Views:	1
Size:	131.1 KB
ID:	1654771

        Comment

        Working...
        X