confint() for difference of effect levels [🇷 for BE/BA]

posted by ElMaestro  – Denmark, 2017-08-29 16:45 (2718 d 17:01 ago) – Posting: # 17751
Views: 8,691

Hi d_labes,

thanks for your input.

❝ No. confint() gives you confidence intervals for the model parameters only.


Except when we fit the model with intercept, treatment first. Now the SE for whichever treatment is not the intercept, is the SE of the difference (and the other treatment effect is unestimable because of the intercept). The funny logic of linear models. I need to get my head around that aspect.


❝ Or if you prefer using such beasts like "least square means" have a look at package lsmeans.


When we look at treatment effect differences it does not matter how we define the effects by contrast coding. I don't absolutely need an LSMean; I can do with two model effects calcuated in any valid way as long as I can extract their difference.

❝ BTW: All-at-once is not the preferred method of a big regulatory body.


Yes, I have heard of that big regulatory body. But there is a problem. All-at-once is the preferred method of another big regulatory body.:-D:-D:-D:-D


❝ Pairwise lm objects (throwing away the data under those treatments not under consideration) will give you the opportunity to use confint() to derive CI's.

So if we are not considering the big agency, but the other big agency :-), and we are not throwing away treatments when fitting models, how would you approach the CI derivation in R?

Many thanks.

Pass or fail!
ElMaestro

Complete thread:

UA Flag
Activity
 Admin contact
23,376 posts in 4,912 threads, 1,664 registered users;
141 visitors (0 registered, 141 guests [including 7 identified bots]).
Forum time: 08:47 CET (Europe/Vienna)

Every system is perfectly designed
to get the results it gets.    Paul B. Batalden

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5