Skip to content

[R--gR] Modelformulae: reference?

6 messages · Steffen Lauritzen, Friedrich Leisch, Achim Zeileis +1 more

#
By the way, does anyone know an authoritative reference for the usual "model
algebra", originating (I presume) with Nelder and Wilkinson and forming the
basis for GLIM model specification and analysis?

Regards
Steffen

--
Steffen L. Lauritzen
Department of Statistics, University of Oxford
1 South Parks Road, Oxford OX1 3TG, United Kingdom
Tel: +44 1865 272877; Fax: +44 1865 272595
email: steffen at stats.ox.ac.uk URL: www.stats.ox.ac.uk/~steffen/
#
> By the way, does anyone know an authoritative reference for the usual "model
  > algebra", originating (I presume) with Nelder and Wilkinson and forming the
  > basis for GLIM model specification and analysis?

I think the original paper is

@Article{flm:Wilkinson+Rogers:1973,
  author       = {G. N. Wilkinson and C. E Rogers},
  title	       = {Symbolic Description of Factorial Models for
                  Analysis of Variance},
  journal      = {Applied Statistics},
  year	       = 1973,
  volume       = 22,
  pages	       = {392--399}
}

but I don't have a copy of the paper, so I don't know what's in there.

best,
fritz
#
Yes, I am sure this is right, thank you. It has no doubt developed a bit
since then, possibly in a partly anarchistic way.

Steffen

----- Original Message ----- 
From: <Friedrich.Leisch at tuwien.ac.at>
To: "Steffen Lauritzen" <steffen at stats.ox.ac.uk>
Cc: "gRlist" <R-sig-gR at stat.math.ethz.ch>
Sent: Friday, August 20, 2004 7:45 AM
Subject: Re: [R--gR] Modelformulae: reference?
"model
forming the
#
> Yes, I am sure this is right, thank you.

You're welcome.

  > It has no doubt developed a bit
  > since then, possibly in a partly anarchistic way.

Yes, certainly. For S, the white book is the key reference.

Best,
Fritz
#
On Fri, 20 Aug 2004 08:45:58 +0200 Friedrich.Leisch at tuwien.ac.at wrote:

            
I put up a pdf copy from JSTOR at
  http://www.R-project.org/gR/papers/

Best,
Z
#
Friedrich.Leisch at tuwien.ac.at writes:
...although much of the "anarchy" is probably of a later date. AFAIR,
the white book model formulas basically copies Genstat notation from
1977 for the lm/glm models (introducing the ~ operator and changing
"." to ":" is about the extent of it, I think), which in turn more or
less *is* Wilkinson/Rogers. Usage in other contexts and things like
the "|" for conditioning in coplot or strata() terms came in later. 

Of course the department copy of the white book is nowhere to be found
so I can't verify.