🔒 This topic is solved and locked.
Need further help from the community? Please
sign in and ask a new question.
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
Posted 12-08-2021 02:51 AM
(806 views)
I have a longitudinal dataset with 1000 observations. Each participant has 2 to 10 timepoints repeated measurement of BMI.
/*id is the unique number for each participant*/ /*here I want to exam the different decline among people who survive and died separately*/ proc mixed data=bmidata PLOTS(MAXPOINTS= 20000) NOCLPRINT; class t id death_status (ref='0'); model bmi = death_status back_timescale death_status*back_timescale age_at_time0 back_timescale_square sex educ /outp=prediction solution ddfm =bw; REPEATED t/ type=vc subject=id r rcorr; random intercept back_timescale /type=VC subject=projid ; run;
I got the results.
So now, I could write the formulation about BMI and these variables.
I am wondering how to calculate the BMI and 95% confidence interval if I valued death_status = 1, back_timescale=-5, age_at_time0 = 75 (around the mean of that timepoint), sex = 1 (female), education = 18 .
Furthermore, I want to compare two BMI values with different death_status but same values for other variables.
1 ACCEPTED SOLUTION
Accepted Solutions
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
2 REPLIES 2
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- RSS Feed
- Permalink
- Report Inappropriate Content
This paper expands on the ideas that Rick shows in his excellent blog post. PLM and the section in the paper on scoring the old fashioned way might be your two best bets.