DOKK / manpages / debian 10 / gbutils / gbtest.1.en
GBTEST(1) User Commands GBTEST(1)

gbtest - Compute statistical tests on data

gbtest [options] <tests list>

Compute statistical tests on data. Each test is specified by a unique name and optional parameters follows, separated by commas. Depending on the test, one, two or three columns of data are expected. '1 sample' tests expect a single columns. Test on 'pairs' expect two columns of equal length. Test on '2 samples' expect two columns of data, possibly of different length. Test on 2+ and 3+ samples expect a varying number of columns. If more columns are provided, the test is repeated for any column in the case of 1 sample test, and any couple of columns in the case of 2 sample tests. In the last case, the output is expressed in matrix format, with estimated statistics on the lower triangle and p-scores (if requested) on the upper. The removal of 'nan' entries is automatic, and is performed consistently with the nature of the test.

Tests typically assume 1, 2 or 3+ samples. Available tests are:

Kolmogorov-Smirnov tests on cumulated data
1 samp Cramer-von Mises tests on cumulated data
1 samp Chi-Sqrd, 1 samp. 2nd column: th. prob.
1 samp Wilcoxon signed-rank test (mode=0)
1 samp Student's T (mean=0)
1 samp Test of randomness: turning points
1 samp Test of randomness: difference sign
1 samp Test of randomness: rank test
pairs Pearson's correlation coefficient
pairs Spearman's Rho rank correlation
pairs Kendall's Tau correlation
pairs Chi-Sqrd, 2 samples
pairs Student's T with paired samples
pairs Wilcoxon test on paired samples
2 samp Kolmogorov-Smirnov test
2 samp Student's T with same variances
2 samp Student's T with different variances
2 samp F-Test for different variances
2 samp Wilcoxon-Mann-Whitney U
2 samp Fligner-Policello standardized U^
2+ samp Levene equality of variances using means
2+ samp Levene equality of variances using medians
3+ samp Kruscal-Wallis test on 3+ samples
3+ samp Multi-columns contingency table analysis

compute associated significance
input data are already sorted in ascending order
specify the input fields separators (default " \t")
set the output format (default '%12.6e')
verbosity: 0 none, 1 headings, 2+ description (default 0)
this help

compute the Fligner-Policello test on the columns of 'file', considering all possible pairings.
compute the Kruscal-Wallis test and its p-score on the first three columns of data in 'file'

Written by Giulio Bottazzi

Report bugs to <>

Package home page <>

Copyright © 2001-2018 Giulio Bottazzi This program is free software; you can redistribute it and/or modify it under the terms of the GNU General Public License (version 2) as published by the Free Software Foundation;

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

March 2018 gbtest 5.7.1