Is there a way to efficiently identify and input missing data for repeated measures data?

SheilaSR — Tue, 23 May 2017 17:10:03 GMT

Hi, I am a student using SAS studio to analyze a set of repeated measures data. The beginning of my dataset looks like this:

subject trt sex age visit time visual
1 Active F 71 1 0 59
1 Active F 71 2 4 55
1 Active F 71 3 12 45
2 Active F 59 1 0 65
2 Active F 59 2 4 70
2 Active F 59 3 12 65
2 Active F 59 4 24 65
2 Active F 59 5 52 55
3 Placebo F 73 1 0 40
3 Placebo F 73 2 4 40
3 Placebo F 73 3 12 37
3 Placebo F 73 4 24 17

In this dataset, "visual" is the outcome measure/response variable. Each subject had the opportunity to have the variable "visual" evaluated 5 times-i.e. at visits 1-5. However, not all subjects made it to all 5 visits. For example, as labeled in red above, subject 1 only had 3 visits, and subject 3 had 4 visits. Subject 2 had all 5.

I would like to enter missing data lines for the missing data. For example:

3 Placebo F 73 5 52 . to indicate that this data is missing.

I would like to do a nonlinear mixed models analysis of this repeated measures data using proc mixed or proc nlmixed.

My questions are:

1. Is there an efficient way in SAS to identify which subjects do not have all 5 visits and then enter the missing data?

2. If data is missing, do I need to enter it into the dataset as I'm describing? Or can I do my analysis without adding these missing rows? I have the impression from the attached SAS guide pages 5-8, that maybe I don't need to add in the missing data lines if I specify the repeated statement appropriately.

I was planning to use the answer from this question: https://communities.sas.com/t5/SAS-Procedures/Adding-extra-data-to-existing-dataset/td-p/117310 to add my missing data lines (or just to enter them in manually), but I was wondering if there is a better way to identify which subjects are missing visit data, rather than going through the 1101 observations in the dataset manually.

Thank you in advance for your help!

Sheila

Re: Is there a way to efficiently identify and input missing data for repeated measures data?

ballardw — Tue, 23 May 2017 17:11:22 GMT

Assuming your data is grouped by subject and sorted by visit as shown this should create the additional records you requested.

data want;
   set have;
   by subject notsorted;
   array x (5) _temporary_ (0, 4, 12, 24,52);
   output;
   if last.subject and visit ne 5 then do;
      call missing(visual);
      do visit=(visit+1) to 5;
         time= x[visit];
         output;
      end;
   end;
end;

Assumes that your time and visit are actually numeric.

Re: Is there a way to efficiently identify and input missing data for repeated measures data?

SheilaSR — Tue, 23 May 2017 17:33:36 GMT

Thank you for your help ballardw! I really appreciate your help! The code worked great and makes sense. Thank you!

topic Is there a way to efficiently identify and input missing data for repeated measures data? in SAS Studio

Is there a way to efficiently identify and input missing data for repeated measures data?

Re: Is there a way to efficiently identify and input missing data for repeated measures data?

Re: Is there a way to efficiently identify and input missing data for repeated measures data?