nscore

Process Help

NSCORE - transform data to a normal distribution

Process Name	Menu Path	Link to Command Table
NSCORE	Estimate ribbon \| Conditional Simulation \| Transform - to Normal	Click here

Introduction
Files, Fields and Parameters
Notes
Example
Errors

Introduction

This is a Superprocess and running it may have an effect on other Datamine files in the project. More...

This process transforms a set of data to a normal distribution. It is based on the GSLIB code for normal score transformation (nscore).

How to use

For details of the method refer to:

GSLIB. Geostatistical Software Library and User's Guide by Clayton V.Deutsch and Andre G.Journel, published by Oxford University Press, Second Edition, 1998. ISBN 0-19-510015-8.

There are some differences between Datamine applications and GSLIB conventions. However, unless otherwise stated, Datamine conventions are used. Where required these are converted to GSLIB conventions by the process.

If a REFDIST file is not specified, then NSCORE transforms the data to a normal distribution using the standard Gaussian cumulative density function method. If a REFDIST file is specified then NSCORE transforms the data to a distribution defined by REFDIST.

The SGSIM process includes an option that carries out a normal transformation prior to the simulation. However it is useful to have the separate process NSCORE to do just the transformation, because the input variogram model for SGSIM should be calculated from the normal data.

Program Limits

Maximum number of: Value

samples in input IN file 500000

Project Folder Name

The following limitations currently apply to the folder name of the project:

A maximum of 48 characters in the full path name of the project folder, for example c:\database\consim\project1 must not exceed 80 characters.
The full path name of the project folder must not include spaces, for example c:\database\consim\my project1 is not allowed.

System Files

_nsclog.txt Log file. Only useful if there is a problem.
_nsc_*.txt Temporary system files. These will be deleted if the process terminates cleanly.
_sp*.dm Temporary Datamine files. These will be deleted if the process terminates cleanly.

All files matching the template _nsc_*.txt and _sp*.dm will be deleted as the process terminates. Therefore you should not use any of these file names.

Files, Fields and Parameters

Input Files

Name	Description	I/O Status	Required	Type
IN	Input sample data file. This must include the grade field GRADE and may also include the declustering weight field DCWGT .	Input	Yes	Table
REFDIST	Optional input reference distribution to define required transformation. The file must include the field REFGRADE , to define the distribution, and may also include the field REFWGT to define declustering weights.	Input	No	Table

Output Files

Name	I/O Status	Required	Type	Description
OUT	Output	Yes	Block Model	Output file containing transformed points. This contains the same data as the IN file, but with the added transformed data field NSCORE.
TRANDIST	Output	No	Table	Output file for the transformation table. This will contain the grade field GRADE from the IN sample file and the field TRANDATA giving the transformed value. The file will be sorted by GRADE .
STAT_TBL	Output	No	Table	Output statistics table. This provides statistics for both the input, untransformed, sample data, as well as the output, transformed, values.

Fields

Name	Description	Source	Required	Type	Default
GRADE	Field in the input sample file IN defining the grade to be transformed.	IN	Yes	Numeric	Undefined
DCWGT	Optional declustering weights field in the IN file.	IN	No	Numeric	Undefined
REFGRADE	Reference grade field, defining the reference distribution in the REFDIST file.	REFDIST	No	Numeric	Undefined
REFWGT	Optional reference weight field, defining declustering weights in the REFDIST file.	REFDIST	No	Numeric	Undefined

Parameters

Name	Description	Required	Default	Range	Values
MINGRADE	Minimum GRADE value input from the IN sample file.	No	0	Undefined	Undefined
MAXGRADE	Maximum GRADE value input from the IN sample file.	No	Undefined	Undefined	Undefined

Notes

No additional notes.

Example

!NSCORE

&IN(dcisaacs),&OUT(nscore1),&TRANDIST(nsctran1),

*GRADE(V),

@MINGRADE=0.0,@MAXGRADE=1700.0

Output Window

NSCORE

Transforming data ....
       1 file(s) copied.
Processing results ....
FORMAT    TIME >11:29:43

Statistics for IN file dcisaacs and OUT file nscore1

=============================================================
File Field     Number  Minimum   Maximum      Mean   Variance
=============================================================
  IN  V            194    0.000  1259.900   294.638  66614.340
  OUT NSCORE       194   -2.749     2.608     0.000      0.996
-------------------------------------------------------------

The following files have been created:
  nscore1: file with 195 records
  nsctran1: transformed distribution file with 194 records

Error and Warning Messages

Message	Description	Solution

The command NSCORE is based on the GSLIB program NSCORE.

The programs in GSLIB are distributed in the hope that they will be useful, but WITHOUT ANY WARRANTY. No author or distributor accepts responsibility to anyone for the consequences of using them or for whether they serve any particular purpose or work at all, unless he says so in writing. Everyone is granted permission to copy, modify and redistribute the programs in GSLIB, but only under the condition that this notice and the above copyright notice remain intact.