automatic exploration of all possible loglinear models?

Thank you all. Warnings well-taken.

My question arose in the context of my "Applied Categorical Data
Analysis" course. We're mainly using SAS. Recent homework questions have
been of the form, "The file provides data on 4 variables A, B, C, and D.
Find a good-fitting model . . . [more questions about interpreting the
model we come up with] . . . "  One of my fellow students asked if SAS
has some way to automate the process (not the interpretation, but at
least fitting the many possible models and providing output for us to
look at.) Apparently SAS does not. I was curious if R did.

So in this case, it is a "required ritual."

We've gone over the dubious worth of the P-values in previous class
discussions.

--Chris
Christopher W. Ryan, MD
SUNY Upstate Medical University Clinical Campus at Binghamton
40 Arch Street, Johnson City, NY  13790
cryanatbinghamtondotedu

"If you want to build a ship, don't drum up the men to gather wood,
divide the work and give orders. Instead, teach them to yearn for the
vast and endless sea."  [Antoine de St. Exupery]

Dieter Menne wrote:

Christopher W. Ryan <cryan <at> binghamton.edu> writes:

Is there a way to automate fitting and assessing loglinear models for
several nominal variables . . . something akin to step or drop1 or add1
for linear or logistic regression?

Not strictly for loglinear, but glm works with stepAIC. Make sure that 
in the field you are working this approach is an accepted ritual.

Dieter

There is also the package formerly known as "dRedging":
http://r-forge.r-project.org/projects/mumin/