topic Missing Data 1 in Statistical Procedures

Missing Data 1

owis — Sat, 13 Aug 2011 16:25:05 GMT

Hi Everyone,

I posted this post before but i added the data set here.

i am working on missing data treatment; i used "proc mi" to do imputation of missing data.

i used this code:

proc mi data = time1 nimpute = 5 seed = 4321567 out=itime1;

run;

But i got these warning messages:

(WARNING: All observed values are identical for variable surveyId. This variable will be excluded from

the analysis.

WARNING: A covariance matrix computed in the EM process is singular. The linearly dependent variables

for the observed data are excluded from the likelihood function. This may result in an

unexpected change in the likelihood between iterations prior to the final convergence.

WARNING: The EM algorithm (MLE) fails to converge after 200 iterations. You can increase the number

of iterations (MAXITER= option) or increase the value of the convergence criterion

(CONVERGE= option).

WARNING: The EM algorithm (posterior mode) fails to converge after 200 iterations. You can increase

the number of iterations (MAXITER= option) or increase the value of the convergence

criterion (CONVERGE= option).

WARNING: The EM algorithm (posterior mode) fails to converge after 200 iterations. You can increase

the number of iterations (MAXITER= option) or increase the value of the convergence

criterion (CONVERGE= option).

WARNING: The posterior covariance matrix is singular. Imputed values for some variables may be fixed.)

I dont know where is the problem. so could anyone help me in that.

Thanks

Owis

Missing Data 1

art297 — Sat, 13 Aug 2011 19:58:36 GMT

I will leave your question to someone who knows more about the method than I do. Some of the warnings were obvious, as you have some variables that are missing for everyone, some variables have the same value for everyone, and some groups of variables have all missing values on virtually every record.

FWIW, I was able to run the proc successfully, separately, for the following groups of variables:

AQLQ1-AQLQ32

_1SEC0--_1SEC6

ACT1-ACT5

SF361--sf36order

AQLQorder--_0secorder

SF361S--SF363jS

GENDER--AQLQ_SA

but got the same warnings, as you did, if I tried to combine those groups.

I look forward to seeing feedback from someone who is an expect on the topic.

Missing Data 1

owis — Sat, 13 Aug 2011 22:30:10 GMT

Thanks for your intersting.

so after that can i combine them again?

Missing Data 1

art297 — Sat, 13 Aug 2011 22:41:17 GMT

I would think so, but would rather hear from someone who understands the method as I REALLY don't know.