Statistics::LinearRegression - simple linear regression
Other
Switch branches/tags
Nothing to show
Permalink
Failed to load latest commit information.
lib/Statistics Add license field to META file Apr 24, 2017
t 1.1.0 changed export rules Nov 19, 2016
.gitignore Add license field to META file Apr 24, 2017
.travis.yml travis and meta updates Nov 17, 2016
LICENSE 'First' Nov 17, 2016
META6.json
README.md 1.1.0 changed export rules Nov 19, 2016

README.md

Build Status

NAME

Statistics::LinearRegression - simple linear regression

SYNOPSIS

Gather some data

my @arguments = 1,2,3;
my @values = 3,2,1;

Build model and predict value for some x using object

use Statistics::LinearRegression;
my $x = 15;
my $y = my LR.new(@arguments, @values).at($x);

If you prefer bare functions, use :ALL

use Statistics::LinearRegression :ALL;
my ($slope, $intercept) = get-parameters(@arguments, @values);
my $x = 15;
my $y = value-at($x, $slope, $intercept);

DESCRIPTION

LinearRegression finds slope and intercept parameters of linear function by minimizing mean square error.

Value at y is calculated using y = slope × x + intercept

TODO

  • R^2 and p-value calculation

  • support for other objective functions

CHANGES

  • 1.1.0 LR class exported by default, bare subroutines need :ALL

AUTHOR

Paweł Szulc pawel_szulc@onet.pl

COPYRIGHT AND LICENSE

Copyright 2016 Paweł Szulc

This library is free software; you can redistribute it and/or modify it under the Artistic License 2.0.