GSLIB Help Page: BICALIB
Whenever soft data are considered in an indicator formalism, it is
necessary to calculate the soft indicator values (the prior
distributions). Also, when considering the Markov-Bayes option in the
indicator simulation program sisim, the B(z) calibration
parameters are called for.
The program bicalib computes the prior distributions and the
B(z) calibration parameters. These prior distributions are
written to one output file and the B(z) parameters are written
to a reporting file from which they must be transferred to the
sisim parameter file.
Often in practice the sparse sample size does not provide enough
resolution for the scattergram of primary vs. secondary variables.
Program scatsmth could be used to
smooth and add resolution to the sample scattergram. Instead of the
original data pairs (size n the input data file datafl to
program bicalib could contain the smoothed sample scattergram.
datafl: a data file in simplified Geo-EAS format containing
the secondary data values (the output file will contain the
corresponding prior distributions for the primary variable).
icolsec: the column number for the secondary variable.
calibfl: a data file in simplified Geo-EAS format containing
pairs of primary and secondary data values (the calibration data).
ivru, ivrv and iwt: the column numbers for the primary
and secondary variables and declustering weight / bivariate frequency.
tmin and tmax: all values strictly less than tmin
and strictly greater than tmax are ignored.
outfl: an output file that contains the prior distributions
corresponding to all input secondary data.
repfl: an output file that contains a report of the pairs
retained, some summary statistics, a bivariate distribution table, the
prior distributions, and the B(z) parameter values.
ncutu: the number of thresholds applied to the primary variable.
cutu(i),i=1,ncutu: the threshold values applied to the primary
ncutv: the number of thresholds applied to the secondary
cutv(i),i=1,ncutv: the threshold values applied to the secondary