Programming the statistical procedures from SAS

weight in proc mixed

Reply
Regular Contributor
Posts: 204

weight in proc mixed

Hi all.

Please correct me if I'm totally off the right track.
I'm interested in finding out which task contributes to higher log_y and because an individual might do more than one task during each measuring Y round, so I thought I could use the weight statement. So if an individual worked longer in task Z than task Y, then task Z would give more weight to contributing Y. But I'm not sure if this is how the code should be like.
Your help is really appreciated.

proc mixed data=tmp2 covtest method=reml cl;
class round Subject task;
model log_y=task/solution;
random int / subject=Subject;
lsmeans task/ pdiff cl;
weight duration;
run;

Data structure:
id plant round log_y task duration
1 A 1 10 Z 5
1 A 1 10 Y 1
1 A 2 30 X 7
1 A 3 25 Z 7
. . . . . .
. . . . . .
. . . . . .
Ask a Question
Discussion stats
  • 0 replies
  • 54 views
  • 0 likes
  • 1 in conversation