#!/usr/local/bin/perl -w

use Statistics::Regression;
	 
# Create regression object
my $reg = Statistics::Regression->new("sample regression", [ "const",
	"X1", "X2", "X3" ] );

# Add data points
$reg->include( 2.0, [ 1.0, 3.0, 1,  -1.0 ] );
$reg->include( 1.0, [ 1.0, 5.0, 0, 2.0 ] );
$reg->include( 20.0, [ 1.0, 31.0, 2,  0.0 ] );
$reg->include( 15.0, [ 1.0, 11.0, 2,  2.0 ] );
$reg->include( 3.0, [ 1.0, 10.0, 1, 1.0 ] );	 
$reg->include( 1.0, [ 1.0, 4.0, 0, 1.0 ] );
$reg->include(100.0, [ 1.0, 50.0, 3, 2.0 ] );

# Finally, print the result
$reg->print();

__END__

Este módulo contiene la versión Perl de 'el código' a usar para realizar
  regresiones múltiples (aunque sea antiguo). It's a must.
  Opcionalmente, se puede pasar al include un peso al final por cada
  observación.

Ejemplo de output:

  ****************************************************************
  Regression 'sample regression'
  ****************************************************************
  Name                   Theta          StdErr     T-stat
  [0='const']         -15.5753         12.1800      -1.28
  [1='X1']              1.6129          0.8070       2.00
  [2='X2']              3.3954         12.5637       0.27
  [3='X3']              5.2290          6.5470       0.80
  
  R^2= 0.877, N= 7, K= 4
  ****************************************************************