.\" It was generated by help2man 1.48.3 and
.\" slightly modified by hand by Jerome Benoit <calculus@debian.org>
.\" on behalf of the Debian Science Team.
.TH PLFIT "1" "July 2021" "plfit" "User Commands"
.SH NAME
plfit \- fits power-law distributions to empirical data
.SH SYNOPSIS
.B plfit
[\fI\,OPTIONS\/\fR] [\fI\,infile \/\fR...]
.SH DESCRIPTION
Reads data points from each given input file and fits a power\-law
distribution to them, one by one, according to the method of
Clauset, Shalizi and Newman. If no input files are given, the
standard input will be processed.
.PP
This implementation uses the L\-BFGS optimization method to find
the optimal alpha for a given xmin in the discrete case. If you
want to use the legacy brute\-force approach originally published
in the above paper, use the \fB\-a\fR switch.
.SH OPTIONS
.TP
\fB\-h\fR
shows this help message
.TP
\fB\-v\fR
shows version information
.TP
\fB\-a\fR RANGE
use legacy brute\-force search for the optimal alpha
when a discrete power\-law distribution is fitted.
RANGE must be in MIN:STEP:MAX format, the default
is 1.5:0.01:3.5.
.TP
\fB\-b\fR
brief (but easily parseable) output format
.TP
\fB\-c\fR
force continuous fitting even when every sample
is an integer
.TP
\fB\-D\fR VALUE
divide each sample in the input data by VALUE to prevent
underflows when fitting discrete power\-law distribution
.TP
\fB\-e\fR EPS
try to provide a p\-value with a precision of EPS when
the p\-value is calculated using the exact method. The
default is 0.01.
.TP
\fB\-f\fR
use finite\-size correction
.TP
\fB\-m\fR XMIN
use XMIN as the minimum value for x instead of searching
for the optimal value
.TP
\fB\-M\fR
print the first four central moments (i.e. mean, variance,
skewness and kurtosis) of the input data to help
assessing the shape of the pdf it may have come from.
.TP
\fB\-p\fR METHOD
use METHOD to calculate the p\-value. Must be one of
skip, approximate or exact. Default is skip.
.TP
\fB\-s\fR SEED
use SEED to seed the random number generator