*********************************************************************************
quad.sas is a sample program for illustration of the Adaptive quadrature method 
used in group testing mixed effect models
Author: Peng Chen
Date: 9-6-2009
*********************************************************************************
1.The first part of the proram simualtes a data set to illustrate the data 
structure needed by SAS proc nlmixed. Practical data set has to be prepared 
in a similar format. That is, each record of the SAS data set has to contain all
information within a group of subjects, including subject covariate vectors, group testing result, site number and group (within site) number.

2.The second part illustrate the use of SAS PROC NLMIXED. Please note that this procedure allows users to embed some data handling code, which can be used to fit group testing 
random effect models.


3.Although the group sizes are the same in the sample data set, they do not have to be. It is easy to see that as long as the group size variable is appropriate, no modification in the program is necessary to accommodate unequal group size situation.

4.This program can also be extended to accormmodate multiple random effects situations.
*********************************************************************************
Note:For simplicity, diagnostic tests are assumed to be perfect. Only minumum changes in the specification of PROC NLMIXED are needed to accormmodate imperfect testing situation.