BookmarkSubscribeRSS Feed
Melk
Lapis Lazuli | Level 10

I am running a mixed model for a continuous measure with years as a continuous independent variable in the study. I also have treatment (1 or 0).

 

Here is my model:

 


proc mixed data=dat;
    class id tx (ref='0');
    model y=years|tx /s cl noint e;
    random intercept years/type=un sub=id;
        estimate 'Control' years 1 years*tx 0 1/cl;
        estimate 'Treatment' years 1 years*tx 1 0/cl;
run;

 

Here is the class level information:SAS Output

tx
21 0

But, my estimates are turning up opposite of what I expect. Are my estimate statements correct or can there be any other glitches I should be aware of?

1 REPLY 1
pau13rown
Lapis Lazuli | Level 10

specify a value for year, i'm guessing 1 is not a legit value. Maybe this helps: https://communities.sas.com/t5/SAS-Statistical-Procedures/How-to-write-estimate-statements-for-conti...

Ready to join fellow brilliant minds for the SAS Hackathon?

Build your skills. Make connections. Enjoy creative freedom. Maybe change the world. Registration is now open through August 30th. Visit the SAS Hackathon homepage.

Register today!
What is ANOVA?

ANOVA, or Analysis Of Variance, is used to compare the averages or means of two or more populations to better understand how they differ. Watch this tutorial for more.

Find more tutorials on the SAS Users YouTube channel.

Discussion stats
  • 1 reply
  • 1312 views
  • 0 likes
  • 2 in conversation