Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

 Function: tree Description: generates from a binary tree an output for plotting.

 Usage: g = tree (t, cut {, opt}) Input: t n x 2 matrix produced from agglom cut scalar the cut-off value for the tree opt text matrix of optional parameters Output: g.points m x 2 matrix for drawing trees via show g.index m x 1 matrix of indices

Note:
At the Cluster analysis tutorial you can find the detailed description of contents.

The 'cut' value gives the level for the 'NaN' values. If

the cut value is larger than 1.5 only the subtree to 1.5

will drawn. The internal used comparison operator is

SMALLER. If the tree is going up the optional parameter

GREATER can be used for cutting. Try the other optional

parameters by yourself:

LEFT (default) the branches of the tree are

left centered

CENTER centered

RIGHT right centered

YAXIS (default) the branches are parallel to the

y-axis

XAXIS x-axis

SMALLER (default) the cut-value is checked with

tree-value smaller as cut

GREATER tree-value greater as cut

SQUARE (default) the tree is drawn with

open rectangles

TRIANGLE open triangles

Example:
```proc()=main()
; load the swiss banknote data
; compute the euclidean distance between banknotes
i=0
d=0.*matrix(rows(x),rows(x))
while(i.<cols(x))
i = i+1
d = d+(x[,i] - x[,i]')^2
endo
d = sqrt(d)
; use the WARD method to cluster the data
t = agglom(d, "WARD", 3)
; generate now the dendrogram for drawing
g = tree(t.g, 0)
g=g.points; build lines
l = 5.*(1:rows(g)/5) +(0:4)' - 4
; create a display for drawing
d = createdisplay(1,1)
; show the dendrogram
show(d, 1, 1, g)
endp
;
main()

```
Result:
```shows as result the dendogram for all 200 swiss banknote data
```

(C) MD*TECH Method and Data Technologies, 05.02.2006