I'd like to try Barlett's approach to multiple imputation using SMC-FCS (Substantive Model Compatible with Fully Conditional Specification). This exists as a package in R, but I have most of my data management work already done in SAS. Is there a way to do this within proc MI/MIANALYZE?
I don't have a dataset to share because I'm just looking for high-level direction at this point. Bartlett article is here for those who are unfamiliar:
https://pubmed.ncbi.nlm.nih.gov/24525487/
Also, the R package smcfcs documentation is here:
https://cran.r-project.org/web/packages/smcfcs/vignettes/smcfcs-vignette.html
Thank you!
Hello,
I guess you already know that the MI Procedure has a FCS statement.
The FCS statement specifies a multivariate imputation by fully conditional specification methods.
I cannot see the
SMC-FCS (Substantive Model Compatible with Fully Conditional Specification)
however.
If you cannot find a paper or a macro with SMC-FCS in SAS, I am afraid it is not available (unless you code it up yourself).
Good luck,
Koen
Ok thank you. That's too bad. Coding it up myself is a bit daunting -- easier for me to just move my project to R.
Calling @Rick_SAS
Available on demand!
Missed SAS Innovate Las Vegas? Watch all the action for free! View the keynotes, general sessions and 22 breakouts on demand.
ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.
Find more tutorials on the SAS Users YouTube channel.