Message-ID: <Pine.LNX.4.61.0505101421240.12508@gannet.stats>
Date: 2005-05-10T13:23:06Z
From: Brian Ripley
Subject: question about k in step
In-Reply-To: <20050509091037.25f07d90@localhost.localdomain>
On Mon, 9 May 2005, ronggui wrote:
>> ?step
> ....
> 'step' uses 'add1' and 'drop1' repeatedly; it will work for any
> method for which they work, and that is determined by having a
> valid method for 'extractAIC'. When the additive constant can be
> chosen so that AIC is equal to Mallows' Cp, this is done and the
> tables are labelled appropriately.
>
> so my question is :what constant should i choose so i can get Mallows'
> Cp instead of AIC?
k=2, that is AIC. Under some circumstances Cp = AIC+const (as it says),
and then the const is chosen so that AIC (which is only determined up to a
const) is equal to Cp (which is defined exactly, at least by Mallows).
--
Brian D. Ripley, ripley at stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595