partial association model

I'm not familiar with the term "partial association model", and I 
don't have the Sloane and Morgan paper you cite.  Neither Google nor 
RSiteSearch("partial association model") produced anything that looked 
to me like what you were asking.

	  Part of the R culture is that one can fit anything with R;  the only 
question is how.  To determine that and to answer your other question on 
the advantages of log-linear models, I believe you could best approach 
those questions by consulting the references in the help file for "glm". 
  In particular, I recommend you start with Venables and Ripley (2002) 
Modern Applied Statistics with S, 4th ed. (Springer).  It is probably 
the best known book on this issue.  If  you are not already familiar 
with this book, I highly recommend it.  I don't have a copy at my 
fingertips now, but with luck, you may find answers to both your 
questions.  It may not discuss "partial association models" by that 
name, but if you know partial association models, you might be able to 
find something relevant in that book.

	  McCullagh P. and Nelder, J. A. (1989) Generalized Linear Models. 
(London: Chapman and Hall) is the second and substantially enlarged 
edition of the initial defining book on this topic.  Hastie, T. J. and 
Pregibon, D. (1992) "Generalized linear models", Chapter 6 of 
_Statistical Models in S_ eds J. M. Chambers and T. J. Hastie, 
(Wadsworth & Brooks/Cole) may provide useful information not available 
in Venables and Ripley.

	  I doubt if this answered your question, but I hope it at least will 
help you in some way.  Feel free to post another question.  However, I 
think you will help yourself if you read and follow the posting guide! 
"http://www.R-project.org/posting-guide.html".  I suspect that on 
average, posts prepared following this guide get more useful answers 
quicker.

	  spencer graves

partial association model

Thread (2 messages)