Message-ID: <A4E5A0B016B8CB41A485FC629B633CED47A5B6F3C6@GOLD.corp.lgc-group.com>
Date: 2012-10-09T15:29:01Z
From: S Ellison
Subject: why does R stepAIC keep unsignificant variables?
In-Reply-To: <EC1AF01D-CD6F-433F-B9FF-FE0D90509E24@comcast.net>
> > Why are the first 5 variables kept in the stepwise result??
I don't know, for your data set.
However, AIC is in my limited experience less likely to reject a term than rejection based on p-value at 95% confidence. It's not the same criterion, so there's no immediate reason it should give the same result.
And it often - quite correctly - doesn't.
S Ellison
*******************************************************************
This email and any attachments are confidential. Any use...{{dropped:8}}