I have this comment in my paper.
(If the difference in the AIC values is not more than 2, it cannot be claimed that the proposed model has a better fit).
Is it correct that.
I found many papers the AIC values between are less than 1 and authors claim is that the new model is better?
Please I need someone explain it for me
Mohammed H A Abujarad
You can used more than one classification, like Deviance, WAIC. Go to library(loo) in R, to get leave one out cross validation.
0 votes
0 thanks
Wesley Bertoli
The answer provided by corey979 on (https://bit.ly/2YihpqD) can be useful to your discussion.
0 votes
0 thanks
- Badges
- Science topic
- Similar topics
- Papers
More Mundher Abdullah's questions See All
Similar questions and discussions