Generalized estimating equation and data format

Question

I first summarize the analysis of some papers that I want to imitate. The regression looks like:

Y2(subjects) = Y2(friends) + Y1(subjects) + Y1(friends) + Z(subjects)

Where subjects refers to focal people and friends refers to those subjects connect to. Some one may be a subject in a relationship and a friend in another. For this question, just skip the validity of this model.

The idea is to use the outcome variable of friends at time 2, Y2(friends), to explain outcome variable of subjects at that time (Y2(subjects)), controlling for outcomes of both subjects and friends at time 1. Other covariates of subjects may be included, but it's not the main consideration.

Following this model, I think the data should be in the wide format, which looks like:

SubjID FrID Y1Subj Y1Fr Y2Subj Y2Fr Z

The authors apply the generalized estimating equation in their analysis. However, I don't know how to perform GEE with the wide data format (let's say PROC GENMOD in SAS), if that's the case. Can you tell if I'm missing some thing (e.g., reconstruction of data in some way so that GEE can be used)?

Answer 1

PROC GENMOD requires data in long format. Data can be reshaped from wide to long format with PROC TRANSPOSE:

For example:

DATA data_wide; 
  INPUT id cpdvar20 cpdvar21 cpdvar22 cpdvar23 ; 
CARDS; 
1 10 13 21 23
2 11 19 18 18
3 12 18 24 27
; 
RUN; 

PROC TRANSPOSE DATA = data_wide out = data_long;
   BY id;
RUN;

/* Add column for year */
DATA data_long;
  SET data_long (rename = (col1=cpdvar));
  year = INPUT(substr(_name_, 7), 5.);
  DROP _name_;
RUN; 

PROC PRINT DATA = data_wide;
RUN;

proc print DATA = data_long;
RUN;

Obs id  cpdvar20 cpdvar21 cpdvar22 cpdvar23
1   1   10       13       21       23
2   2   11       19       18       18
3   3   12       18       24       27

Obs id  cpdvar  year
1   1   10      20
2   1   13      21
3   1   21      22
4   1   23      23
5   2   11      20
6   2   19      21
7   2   18      22
8   2   18      23
9   3   12      20
10  3   18      21
11  3   24      22
12  3   27      23
```