[R] Minimizing both AIC and df in model selection
Christoph.Scherber at agr.uni-goettingen.de
Wed Oct 1 16:22:19 CEST 2008
Dear R users,
Suppose I have a set of 10 candidate models, and these all differ by a delta-AIC of 5.
Would it be sensible, then, to choose the best-fit model as the one that has *both* minimal AIC
*and* minimal number of d.f.?
Are there any references for such an approach?
Many thanks for your help!
Dr. rer.nat. Christoph Scherber
University of Goettingen
phone +49 (0)551 39 8807
fax +49 (0)551 39 8806
More information about the R-help