SASian potpourri [Study As­sess­ment]

posted by d_labes  – Berlin, Germany, 2016-05-24 12:02  – Posting: # 16358
Views: 17,463

(edited by d_labes on 2016-05-24 12:49)

Dear All!

» ... Would be great if one of our SASians could jump in.

Here we go:

Software               Estimate       90% CI
NCSS               B0 1.9414    1.4849    2.3978   
                   B1 0.7617    0.6659    0.8576   
Phoenix/WinNonlin  B0 1.9413858 1.4967592 2.3860125
                   B1 0.7617406 0.6695783 0.8539029
R lmerTest         B0 1.9413858 1.4967584 2.3860132
                   B1 0.7617406 0.6695768 0.8539044
SAS (ML)           B0 1.9413856 1.5174574 2.3675030
                   B1 0.7614780 0.6738351 0.8491210
SAS (REML)         B0 1.9413856 1.4807360 2.4020351
                   B1 0.7617407 0.6664695 0.8570118
Smith et al.       B0 1.94      1.54      2.35     
                   B1 0.7615    0.679     0.844    
SAS (ML/satterth)  B0 1.9425    1.5391    2.3458   
                   B1 0.7615    0.6789    0.8441   

Smith et al. used ML (usual maximum likelihood) as estimation method! Degrees of freedom method is per default "containment" (7 for the intercept, 5 for the regression slope). Astonishing enough I couldn't reproduce their results w.r.t. to the 90% CIs to a sufficient degree of accuracy with this ddfm method, at least sufficient for me.
Choosing ddfm=SATTERTHWAITE gives the desired results.

SAS code:
Proc mixed data=dp method=ML;
  class subject;
  model lnCmax=lnd /s cl alpha=0.1 ddfm=SAT;
  random intercept /subject=subject;
run;



edit: the missing REML+SATTERTH
SAS (REML/satterth) B0 1.9414    1.4968    2.3860     
                    B1 0.7617    0.6696    0.8539   

Bingo! Same as R lmer() and Phoenix/WinNonlin, at least with 4 decimals, SAS default output format. Too lazy to tease out more numbers :cool:.

Regards,

Detlew

Complete thread:

Activity
 Mix view
Bioequivalence and Bioavailability Forum |  Admin contact
19,889 posts in 4,215 threads, 1,364 registered users;
online 6 (0 registered, 6 guests [including 3 identified bots]).
Forum time (Europe/Vienna): 08:24 CEST

I have no opinion about ‘incurred samples’ –
an expression which has no easily understandable
meaning for me in the English language.    Nick Holford

The BIOEQUIVALENCE / BIOAVAILABILITY FORUM is hosted by
BEBAC Ing. Helmut Schütz
HTML5