I need to calculate an index of predictive association, similar to Goodman and
Kuskal's (1954) for a 2x2x2 array. I have two questions:
1) Is there an easy way to do this in R? (Of course, the answer to this kind of
question is nearly always ``yes.'' Perhaps I should have asked, ``How
can I do
this in R?'')
2) Is it reasonable to try this with a 3-dimensional array? The way I thought
of doing it (say, the variables are x, y, and z, with z being the
outcome) involves calculating the probability of z_i conditional on knowing
both x and y. This leads to some funny stuff, which makes me doubt if I'm
doing
it right. Perhaps I should condition on only one of x and y at a time?
Any help would be appreciated.
______________________________________________________________________
Stuart Luppescu -=-=- University of Chicago
$(B:MJ8$HCRF`H~$NIc(B -=-=- s-luppescu at uchicago.edu
http://www.consortium-chicago.org/people/sl/sl.html
PGP Public Key: www.consortium-chicago.org/people/sl/pubkey.asc
ICQ #21172047 AIM: psycho7070
tax office, n.:
Den of inequity.>> Sent on 11-Aug-2000 at 14:36:43 with xfmail
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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 at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
