R help - Jul 2012 - Definition of AIC (Akaike information criterion) for normal error models

Dear R users (r-help@r-project.org),

   The definition of  AIC (Akaike information criterion)
for normal error models has just been changed.
   Please refer to the paper below on this matter. Eq.(22) is
the new definition. The essential part is RSS(n+q+1)/(n-q-3);
it is close to GCV.  The paper is temporarily available at
the "Papers In Press" place.

Kunio Takezawa(2012): A Revision of AIC for Normal Error Models
Open Journal of Statistics, Vol.2. No.3
http://www.scirp.org/journal/ojs/

**Out now!**
Guidebook to R Graphics Using Microsoft Windows
ISBN: 978-1-118-02639-7
Click here to read more and to buy your print copy
http://as.wiley.com/WileyCDA/WileyTitle/productCd-111802639X.html

***** http://cse.naro.affrc.go.jp/takezawa/intro.html *****

	[[alternative HTML version deleted]]

On 06/07/2012 09:14, Kunio takezawa wrote:
> Dear R users (r-help at r-project.org),
>
>     The definition of  AIC (Akaike information criterion)
> for normal error models has just been changed.
Excuse me, who empowered you to re-write history?  Maybe Dr Akaike could 
have changed his definition, but he is no longer with us to do so.

And 'AIC' was not an abbreviation of 'Akaike information
criterion'
according to Dr Akaike ....

I am sure that R will continue to use the definition given by Dr Akaike 
for its AIC() function: if your intention was to try to get that 
changed, this was the wrong list to post to.

-- 
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