gbacorr(1) Compute auto/cross-correlation coefficients


gbacorr [options]


Compute auto/cross-correlation coefficients

If the input is a single columns x_1...X_T, the autocorrelation function c(t) is printed, defined as

c_{t} = 1/(T-t-1) \sum_i (x_i-m) (x_{i+t}-m) /s^2

where m is the sample average and s the standard deviation. With a second column y_1...y_T, the cross-correlation

c_{t} = 1/(T-t-1) \sum_i (x_i-mx) (y_{i+t}-my) /(sx sy)

is printed where mx and my are the average values of the two columns and sx and sy their standard deviations. With -M 1 it is mx=my=0, the is computed accordingly and in the previous formula T-t-1 is replaced by T-t. The range of t is set by option -t. Options

choose the method (default '0'):
auto/cross-correlation with mean removal,
auto/cross-correlation without mean removal
set range of t (default '0,10'), accept negative integers
specify the confidence level in (0,1). Interval ac_low,ac_hi has a probability 1-confidence to contain the true value. With this option the output becomes: lag ac ac_low ac_hi.
specify the input fields separators (default " \t")
this help


gbacorr -t 0,2 'file(1)'
first three a.c. coeff. of the first data column
gbacorr -p 0.05 'file(1:2)' x-corr of the first two columns together with
their 5% confidence intervals


Written by Giulio Bottazzi


Report bugs to <[email protected]>

Package home page <>


Copyright © 2001-2015 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.