Control For Tree Growing

DESCRIPTION:

Returns a list of control parameters for the algorithm used by tree.

USAGE:

tree.control(nobs, mincut=<<see below>>, minsize=<<see below>>,  
      mindev=0.01) 

REQUIRED ARGUMENTS:

nobs
number of observations in data used to grow a tree.

OPTIONAL ARGUMENTS:

mincut
minimum number of observations to include before the first cut on a variable. This can not be greater than half of minsize. The default is the smaller of 5 and trunc(minsize/2).
minsize
node size at which the last split is performed, i.e. minsize=5 means that growing continues if there are at least 5 observations in a node. The default is the larger of 10 and 2*mincut.
mindev
minimum node deviance before growing stops.

VALUE:

a list
mincut
the maximum of the input or default mincut and 1.
minsize
the maximum of the input or default minsize and 2.
mindev
the input or default mindev.
nmax
the ceiling of the smaller of 4*nobs/(minsize-1) and 4/mindev, where minsize is the output minsize.
nobs
the value of the argument nobs.

DETAILS:

To fit a tree perfectly to the data, use minsize=2 and mindev=0.

SEE ALSO:

.