topic Re: Multilevel Imputation in SAS Programming

Multilevel Imputation

mahmood — Thu, 29 Jun 2017 14:48:00 GMT

Hello,

I learned that there is a SAS macro for Mutilevel Imputation called "MMI_IMPUTE". Is there any way that I can get this macro? Would anyone help me please?

Thanks,

Rahid Mahmood

Re: Multilevel Imputation

ballardw — Thu, 29 Jun 2017 15:12:48 GMT

This paper https://support.sas.com/resources/papers/proceedings13/438-2013.pdf indicates a link to source code. I can't confirm that as the site the link in the paper uses is blocked by my organization.

Re: Multilevel Imputation

mahmood — Thu, 29 Jun 2017 15:41:57 GMT

Hi, Thanks. But that website is not active anymore.

Re: Multilevel Imputation

StatsGeek — Thu, 10 Aug 2017 19:27:43 GMT

Sorry about the website confusion. I took down the website when I stopped offering consulting services. As for the macro, I no longer recommend using MMI_IMPUTE. MMI_IMPUTE uses an imputation algorithm called PAN, which was developed a while back by Joseph Schafer. Unfortunately, PAN is a bit outdated, and is less flexible than newer algorithms (e.g., the algorithm can't incorporate random effects between incomplete variables; all incomplete variables are required to be normally distributed). I recommend that you instead use a standalone software package called Blimp. Blimp was written by Brian Keller and Craig Enders at UCLA. Unlike MMI_IMPUTE, Blimp can handle random effects between incomplete variables and can also handle some non-normal incomplete variables (e.g., binary variables). The software and associated documentation are available at http://www.appliedmissingdata.com/multilevel-imputation.html. Blimp is free, and the website contains scripts for using it from SAS.

Re: Multilevel Imputation

halladje — Sat, 22 Jun 2019 13:28:02 GMT

Is there a paper you recommend talking about how PAN and MMI_IMPUTE work? Is there anything published regarding what you said re: outdated?

Also, can MMI_IMPUTE handle 3 level, or only 2-level? I am working on a paper comparing different missing data strategies in multilevel models (including BLIMP) and I am just wondering if I should also include MMI_IMPUTE for comparison purposes, or if something has already been published that I can reference.

Thanks so much!