emmeans for lme when the data is incomplete [🇷 for BE/BA]

posted by mittyri – Russia, 2023-05-20 22:19 (396 d 23:56 ago) – Posting: # 23561
Views: 1,317

Dear Helmut,

let me raise this old but useful post
first of all,

❝ A small step for a man but a giant leap for mankind?

:flower: should go to Detlew, take a look at his proposal
this is pretty the same, I didn't find any differences except some other rules of arguments naming.
I tried it for some models and it works well until uncomplete data is given:

> test(pairs(emmeans(mod, ~ Formulation, mode = "satterth"),
+            reverse = TRUE), delta = log(0.8))

Error in (mth$objs[[1]])(object, trms, xlev, grid, ...) :
  Unable to estimate Satterthwaite parameters

Any ideas? should I contact Russ regarding that?
It works when the mode is asymptotic or containment

Kind regards,

Complete thread:

UA Flag
 Admin contact
23,059 posts in 4,841 threads, 1,647 registered users;
36 visitors (0 registered, 36 guests [including 7 identified bots]).
Forum time: 22:16 CEST (Europe/Vienna)

Friends don’t let friends use Excel for statistics!    Jonathan D. Cryer

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz