 Usage:  cw = wardcont (x, k, l)  
 
 Input:

  x                      n x p matrix of n row points to be clustered (the 
                         elements must be >= 0, with positive marginal 
                         sums) 
                         
  k                      scalar the maximum number of clusters of rows 
                         
  l                      scalar the maximum number of clusters of columns 
                         
 Output:

  cw.y                   n x l matrix correspondence analysis scores of the 
                         row points (l = min(n-1,p-1) 
                         
  cw.z                   p x l matrix: correspondence analysis scores of the 
                         column points 
                         
  cw.r                   n x 1 matrix: Partition of n points into k clusters 
                         
  cw.c                   p x 1 matrix: Partition of p points into l clusters 
                         
--------------------------------------------------------------
(C) MD*TECH Method and Data Technologies, 17.8.2000
