GSLIB Help Page: SCATSMTH

Description:

Creates a smooth bivariate distribution model constrained to marginal
distributions, correlation coefficient, bivariate quantiles, and
smoothmess

Parameters:

datafl: the data file with the raw (perhaps declustered)
paired observations.

icolx, icoly and icolwt: the column location for the
z_1 variable, the z_2 variable and the declustering weight
(0 if none available).

smoothx: the data file with the smoothed z_1 (X)
distribution.

icolvr and icolwt: the column location for the variable
and the weight (probability).

smoothy: the data file with the smoothed z_2 (Y)
distribution.

icolvr and icolwt: the column location for the variable
and the weight (probability).

dbgfl: name for the output file for debugging information.

finalxfl: name for the output file of the final X
distribution  to check reproduction of the marginal distribution

finalyfl: name for the output file of the final Y
distribution  to check reproduction of the marginal distribution

outfl: the output file. This file will contain the N_1
(nx) by N_2 (ny) X Y and probability values.

maxpert, report, omin and seed: after
maxpert x nz perturbations the program is stopped.
After report x nz perturbations the program reports on
the current objective function(s). When the normalized objective
function reaches omin the program is stopped.
The random number seed seed should be a large odd integer.

imarg, icorr, ismth and iquan: flags for whether
closeness to the marginal histograms, closeness to a target correlation
coefficient, smoothness, and closeness to specified
quantiles are considered (1 = yes, 0 = no).

sclmarg, sclcorr, sclsmth and sclquan: user imposed
weights which scale the weights that the program automatically
calculates (default is 1.0 for all weights).

nsmooth: size of the smoothing window.

correlation: target correlation coefficient.

ndqx and ndqy: number of quantiles defined from the data
(evenly spaced cumulative probability values in x and y).

nenvelope: number of points defining the envelope containing
the bivariate distribution (may be set to zero so that all points in
the bivariate distribution are allowable)

x(),y(): points defining the allowable bivariate envelope. This
``envelope'' is a practical feature that restricts the nonzero
bivariate probability values to fall within some userspecified
region.