Proc MI

jojo — Fri, 21 Feb 2025 06:39:27 GMT

anyone used proc MI for longitudinal data in long format before?

data da;

input id$ visitn trt val;

data lines;

001 1 A 32

001 2 A 29

001 3 A 25

001 4 A 30

001 5 A 22

002 1 A .

002 2 A 20

002 3 A .

002 4 A 18

002 5 A 28

003 1 A 15

003 2 A 10

003 3 A .

003 4 A 18

003 5 A 18

004 1 B 15

004 2 B 17

004 3 B 20

004 4 B 18

004 5 B .

005 1 B 16

005 2 B 14

005 3 B 20

005 4 B .

005 5 B .

006 1 B 19

006 2 B 17

006 3 B 20

006 4 B .

006 5 B 18

;

run;

proc mi data=da nimpute=50 out=da1;

mcmc chain=multiple;

var visitn val;

by id;

run;

Re: Proc MI

SAS_Rob — Fri, 21 Feb 2025 22:30:06 GMT

The long format is not usually compatible with performing multiple imputation, thus data restructuring from long to wide or the reverse is often needed for multiple imputation. There is a good discussion in Raghunathan's Missing Data in Practice text (2016) pages 121-126 and in the 2018 paper linked below.

Using SAS for Multiple Imputation and Analysis of Longitudinal Data

topic Re: Proc MI in Statistical Procedures

Proc MI

Re: Proc MI