Homework 21. Due Monday April 21, 2008.
---------------------------------------

In SAS and R: Create an Analysis of Deviance table for 
(i) Logistic Regression, and (ii) Probit Regression 
using all of the regressors in the kyphosis data 
PLUS Age^2.  You can find the data at
http://www.math.umd.edu/~evs/~s798c/Data web-page. 
 
These data are discussed in a handout, which can be found under  
http://www.math.umd.edu/~evs/s770/SASlogs/SASlog3.txt .

As far as I know the most economical way to do the analysis 
of deviance table in SAS is to do your PROC LOGISTIC 
(or GENMOD) regressions using forward or backward selection 
(with a very forgiving criterion for variable inclusion if 
SELECTION=FORWARD and a very restrictive criterion for 
variable retention if SELECTION=BACKWARD.)

For each of your best fitted models (one using logit and 
one probit), discuss the goodness of fit by interpreting 
the cross-tabulation of the Kyphosis response versus the 
quantity   trunc(5*glmfit$fit)/5  .  [Do this part only in R .]