modified function for selection of time points [🇷 for BE/BA]

posted by Aceto81 – Belgium, 2008-10-16 14:09 (6445 d 13:34 ago) – Posting: # 2542
Views: 26,067

❝ Do you think that we need to take the absolute value for AIC (extractAIC) here? :confused: AICs can be negative and the smallest is the best, as far as I can remember. I know that these codes were written by Ace previously. What do you think about this?

return(cbind((nrow(dat)-m+1),abs(extractAIC(lm(log(conc)~time,dat[m:nrow(dat),])))[2],summary(lm(log(conc)~time,dat[m:nrow(dat),]))$adj.r.squared))


Maybe a bit of clarification for the abs() value:
I took the example from a thread, mentioned before, and get the same AIC values, except that they were negative. So that's the reason why I used the abs() value.

Ace

Complete thread:

UA Flag
Activity
 Admin contact
23,653 posts in 4,991 threads, 1,570 registered users;
99 visitors (0 registered, 99 guests [including 16 identified bots]).
Forum time: 03:43 CEST (Europe/Vienna)

In theory, there is no difference between theory and practice.
But, in practice, there is.    Jan L.A. van de Snepscheut

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