Programming the statistical procedures from SAS

What is the 'cutoff' value for Information Criteria like AIC and BIC?

Accepted Solution Solved
Reply
Contributor
Posts: 45
Accepted Solution

What is the 'cutoff' value for Information Criteria like AIC and BIC?

When comparing two IC values, what is the cutoff value?

I have heard that it was 10, and also heard that it was 2...


Accepted Solutions
Solution
‎12-11-2013 03:02 PM
Respected Advisor
Posts: 2,655

Re: What is the 'cutoff' value for Information Criteria like AIC and BIC?

See Burnham and Anderson's Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach (2nd. ed) (2002) from Springer-Verlag--Chapter 6, section 4.5, according to my notes.

Steve Denham

View solution in original post


All Replies
Contributor
Posts: 62

Re: What is the 'cutoff' value for Information Criteria like AIC and BIC?

Hi,

Note : The difference is based on the AIC (BIC) of your desired model with the smallest AIC (BIC) in all your models.

For AIC , Page 30 : http://myweb.uiowa.edu/cavaaugh/ms_lec_2_ho.pdf

For BIC , Page 30 : http://myweb.uiowa.edu/cavaaugh/ms_lec_5_ho.pdf

Respected Advisor
Posts: 2,655

Re: What is the 'cutoff' value for Information Criteria like AIC and BIC?

For AIC, there should be no "cutoff" value, in my opinion.  The smaller value (in SAS parlance) for two models reflects less information loss, as compared to the data generating process/model.  The loss compared to the model with the minimum AIC (AICmin) can be calculated as exp((AIC - AICmin)/2).  Thus even a small difference can indicate substantial information loss, e.g. a difference of 1.386 in AIC would represent a 50% loss.

Steve Denham

Contributor
Posts: 62

Re: What is the 'cutoff' value for Information Criteria like AIC and BIC?

Thank you SteveDenham.

I don't believe in "cutoff" for AIC as you say.But his question motivate me to search internet to find some rules. I don't use these rules. But if someone asks me have you ever heard about "cutoff" for AIC . I referred him to these pretensions. and then if he wants my opinion about using these "cutoff" , I tell him/her your invaluable example. Smiley Happy

Thank you

Contributor
Posts: 45

Re: What is the 'cutoff' value for Information Criteria like AIC and BIC?

SteveDenham and MohammadFayaz, thank you for your input. Do you have a reference for the exp((AIC - AICmin)/2) formula you gave? Thank you Marcio

Solution
‎12-11-2013 03:02 PM
Respected Advisor
Posts: 2,655

Re: What is the 'cutoff' value for Information Criteria like AIC and BIC?

See Burnham and Anderson's Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach (2nd. ed) (2002) from Springer-Verlag--Chapter 6, section 4.5, according to my notes.

Steve Denham

Contributor
Posts: 45

Re: What is the 'cutoff' value for Information Criteria like AIC and BIC?

Thank you I will check. And how about when you have two exactly similar AIC and BIC between a linear and a quadratic model? Which one do we pick?

Respected Advisor
Posts: 2,655

Re: What is the 'cutoff' value for Information Criteria like AIC and BIC?

Exactly similar?  That would be very coincidental.  I'd pick the model that I had the most solid biological reason for selecting, though, given this strange occurrence.

Steve Denham

Contributor
Posts: 45

Re: What is the 'cutoff' value for Information Criteria like AIC and BIC?

Yes, Thank you Very much Steve

🔒 This topic is solved and locked.

Need further help from the community? Please ask a new question.

Discussion stats
  • 8 replies
  • 1928 views
  • 6 likes
  • 3 in conversation