🚀 go-pugleaf

I posted a query about OO statistical software quite a long time ago
(for the benefit of a MSc thesis writer).
The following persons answered (thank you all!):

mivirtan@valt.helsinki.fi (MIKKO VIRTANEN (TIL))
dmurdoch@mast.QueensU.CA (Duncan Murdoch)
Ronald Schoenberg <rons@u.washington.edu>
rbosch@stat.uiowa.edu (Ron Bosch)
sdl@linus.mitre.org (Steven Litvintchouk)
ail8070@tamuts.tamu.edu (Andy I. Liaw)
Kok Hooi Tan <kt14+@andrew.cmu.edu>
adr@ccr.jussieu.fr (Adrian RAFTERY p73343)
Rauno Veijola <veijola@galahad.jyu.fi>
knox@spruce.gsfc.nasa.gov (Robert Knox)
almond@statsci.com (Russell G. Almond)
eb@icnucevx.cnuce.cnr.it (Enrico Badella)

------------------------------------------------------------------------
The following two software packages and their associated
books were mentioned in many answers:

	Luke Tierney (1990): Lisp-Stat - An object oriented environment
		for statistical computing and dynamic graphics
		 Wiley, N Y

Xlisp-stat is available through statlib@stat.cmu.edu.

Chambers, J.M, and T.J. Hastie. 1992. Statistical Models in S.  
   Wadsworth & Brooks,  Pacific Grove, California. (esp. 
   appendix A, by J.M. Chambers)

I apologize if this is irrelavent:  the software package S-plus
from Stat Sci is a OOP stat computing envrionment that is widely
used here for stat research (some use the earlier version, New-S
from AT&T).

------------------------------------------------------------------------
Other literature, products and projects mentioned:

M. Sibuya and R. Shibata (1992) `Data Analysis by using S'  
   Kyouritsu Syuppan, Japan.

I think there's supposed to be a book length treatment of object 
oriented programming in S, coming out soon, but I don't have the
details handy and may be mis-remembering.

And another is xgobi, but I don't remember any bibliographical
information about it.

For a more theoretical discussion, look for a joint paper by R. W. Oldford 
and C. Hurley a few years back (I can't find the reference).  Oldford has a 
couple of papers listed in the 1988 Current Index to Statistics that are 
probably relevant, as does Hurley.

You should also check out the Interface proceedings.

The DINDE project (Wayne Oldford, University of Waterloo) and the
Arizona project (John McDonald, University of Washington) are both
attempts to develop OO statistical systems in LISP.

Is he interested only in literature, or would he also like to know
about languages/software as well?  Is so, I would suggest that M++,
which as you know are C++ array classes, is a serious attempt to
generate an object-oriented scientific programming language.  If he is
interested in looking into it, he could get in touch with me.  I've
designed least squares classes, and optimization classes (which could
be used for maximum likelihood estimation), and I'm writing a paper
for the OON-SKI '93 on these designs.
  (From: Ronald Schoenberg <rons@u.washington.edu>)

Our work on the statistical identification of ice-floes in satellite
images depended crucially on object-oriented programming ideas.
See Banfield and Raftery (1992), "Identification of ice floes ...",
JASA, p.7.
  (From: adr@ccr.jussieu.fr (Adrian RAFTERY p73343))

 ...                                                 Also ElToY, (my
own work) which is a large scale attempt to work in it is available
through statlib.  ElToY contains, among other things, and object
protocol for probability distributions.
 ...
There are a number of research projects here at StatSci aimed at
bringing more of S-PLUS into the object oriented framework.
  (From: almond@statsci.com (Russell G. Almond))

----------------------------------------------------------------------
Markku Sakkinen (sakkinen@jytko.jyu.fi)
       SAKKINEN@FINJYU.bitnet (alternative network address)
Department of Computer Science and Information Systems
University of Jyvaskyla (a's with umlauts)
PL 35
SF-40351 Jyvaskyla (umlauts again)
Finland
----------------------------------------------------------------------