Re: [R] GLM query

About this list Date view Thread view Subject view Author view

Prof Brian Ripley (ripley@stats.ox.ac.uk)
Tue, 21 Dec 1999 09:43:08 +0000 (GMT) 




> From: spoon <spoon@hilbert.maths.utas.edu.au>


> Date: Tue, 21 Dec 1999 19:30:40 +1100 (EST)


> 



[...]


AIC: -70.016


 


> But I get a very different AIC to Lindsey (he gets 254.3).  I've


> played around a bit and have no idea how to get his solution.  Any


> clues?



Here's my guess.  Your AIC is based on a log-likelihood regarding log(GDP)


as the data, and his is regarding GDP as the data. The transformation


of the density adds a term to the log-likelihood which is the same for each 


model.



Test:



> 2*sum(log(gdp$GDP))


[1] 324.4009



seems close enough to the difference, although I would have expected it


to be closer.



-- 
Brian D. Ripley,                  ripley@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 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
















About this list
Date view
Thread view
Subject view
Author view











This archive was generated by hypermail 2.0b3 
on Tue 04 Jan 2000 - 13:34:03 EST