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

posted by Aceto81 – Belgium, 2008-10-16 14:09 (6463 d 19:25 ago) – Posting: # 2542
Views: 26,288

❝ 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
23,655 posts in 4,993 threads, 1,572 registered users;
344 visitors (0 registered, 344 guests [including 18 identified bots]).
Forum time: 09:34 CEST (Europe/Vienna)

The real struggle is not between the right and the left
but between the party of the thoughtful
and the party of the jerks.    Jimmy Wales

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