G::Tools
Statistics
Summary
G::Tools::Statistics - Statistical Methods
Package variables
No package variables defined.
Included modules
Inherit
Exporter
Synopsis
use G::Tools::Statistics;
$mean = mean(@values);
Description
This module contains statistical analysis methods, which is
mostly a simple wrapper around CPAN Statistics:: modules.
Methods
Methods description
Name: cor - calculate the correlation of given two array references
Description:
Calculates correlation coefficient from two array references, using
Pearson's, Spearman's or Kendall's methods.
This is a wrapper around Statistics::Descriptive and Statistics::RankCorrelation modules.
Usage:
$r = cor([@array1], [@array2]);
Options:
-method "pearson", "spearman", or "kendall" (default:pearson)
Method used to calculate the correlation coefficient
-sorted sorted rank coefficient when 1 (default:0)
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20130422-01 changed the default value to Pearson's correlation rather than R^2
20070607-01 initial posting |
Name: cumulative - returns cumulative array of the given array
Description:
Returns the cumulative array of the data.
Usage:
@cumulative = cumulative(@array_of_values);
Options:
None.
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20081124-01 initial posting |
Name: least_squares_fit - calculate the least squares fit of the given array of data
Description:
Performs least squares fit on the data.
When called in array context, this returns an array of four
values ($intercept, $slope, $r, $error), where the linear fit is expressed by
y = $slope * x + $intercept, and $r is the Pearson's linear correlation coefficient,
$error is the root-mean-square error.
When called in scalar context, only $r is returned.
This is a wrapper around Statistics::Descriptive module.
Usage:
($intercept, $slope, $r, $error) = least_squares_fit(@array_of_values);
or
$r = least_squares_fit(@array_of_values);
or
$r = scalar least_squares_fit(@array_of_values);
Options:
None.
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20070607-01 initial posting |
Name: max - get the maximum value of the given array of data
Description:
Returns the maximum value of the data.
This is a wrapper around Statistics::Descriptive module.
Usage:
$max = max(@array_of_values);
Options:
None.
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20070607-01 initial posting |
Name: maxdex - get the index of maximum value of the given array of data
Description:
Returns the index of the maximum value of the data.
This is a wrapper around Statistics::Descriptive module.
Usage:
$maximum_index = maxdex(@array_of_values);
Options:
None.
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20070607-01 initial posting |
Name: mean - calculate the mean of the given array of data
Description:
Returns the mean of the data.
This is a wrapper around Statistics::Descriptive module.
Usage:
$mean = mean(@array_of_values);
Options:
None.
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20070607-01 initial posting |
Name: median - calculate the median of the given array of data
Description:
Returns the median of the data.
This is a wrapper around Statistics::Descriptive module.
Usage:
$median = median(@array_of_values);
Options:
None.
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20070607-01 initial posting |
Name: min - get the minimum value of the given array of data
Description:
Returns the minimum value of the data.
This is a wrapper around Statistics::Descriptive module.
Usage:
$min = min(@array_of_values);
Options:
None.
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20070607-01 initial posting |
Name: mindex - get the index of minimum value of the given array of data
Description:
Returns the index of the minimum value of the data.
This is a wrapper around Statistics::Descriptive module.
Usage:
$minimum_index = mindex(@array_of_values);
Options:
None.
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20070607-01 initial posting |
Name: standard_deviation - calculate the standard deviation of the given array of data
Description:
Returns the standard deviation of the data. Division by n-1 is used.
This is a wrapper around Statistics::Descriptive module.
Usage:
$standard_deviation = standard_deviation(@array_of_values);
Options:
None.
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20070607-01 initial posting |
Name: sum - calculate the sum of the given array of data
Description:
Returns the sum of the data.
This is a wrapper around Statistics::Descriptive module.
Usage:
$sum = sum(@array_of_values);
Options:
None.
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20070607-01 initial posting |
Name: ttest - performs Student's t-test on given two array references
Description:
Performs Student's t-test (independent or pair-wise) on two array references.
Only the p-value is returned in scalar context. If called in array context,
an array of 3 values ($t_value, $p_value, $degree_of_freedom) are returned.
This is a wrapper around Statistics::TTest and Statistics::DependantTTest modules.
Usage:
($t_value, $p_value, $df) = ttest([@array1], [@array2]);
or
$p_value = ttest([@array1], [@array2]);
or
$p_value = scalar ttest([@array1], [@array2]);
Options:
-paired 1 for dependent (paired t-test), 0 for independent (default:0)
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20070607-01 initial posting |
Name: variance - calculate the variance of the given array of data
Description:
Returns the variance of the data. Division by n-1 is used.
This is a wrapper around Statistics::Descriptive module.
Usage:
$variance = variance(@array_of_values);
Options:
None.
Author:
Kazuharu Arakawa (gaou@sfc.keio.ac.jp)
History:
20070607-01 initial posting |
Methods code
sub cor
{ opt_default(method=>"pearson", sorted=>0);
my @args = opt_get(@_);
my $method = opt_val("method");
my $sorted = opt_val("sorted");
if($method eq 'spearman' || $method eq 'kendall'){
require Statistics::RankCorrelation;
my $stat = Statistics::RankCorrelation->new(@args, sorted=>$sorted);
return $stat->spearman if($method eq 'spearman');
return $stat->kendall if($method eq 'kendall');
}else{
require Statistics::Descriptive;
my $stat2 = Statistics::Descriptive::Full->new();
$stat2->add_data(@{$args[0]});
my @result = $stat2->least_squares_fit(@{$args[1]});
return $result[2];
}} | sub cumulative
{ opt_default("mean"=>0);
my @args = opt_get(@_);
my $mean = opt_val("mean");
my @array;
my $cum;
my $ave = $mean ? mean(@args) : 0;
foreach my $val (@args){
$cum += $val - $ave;
push(@array, $cum);
}
return @array;} | sub least_squares_fit
{ require Statistics::Descriptive;
my $stat = Statistics::Descriptive::Full->new();
$stat->add_data(@{$_[0]});
my @result = $stat->least_squares_fit(@{$_[1]});
if(wantarray()){
return @result;
}else{
return $result[2];
}} | sub max
{ require Statistics::Descriptive;
my $stat = Statistics::Descriptive::Full->new();
$stat->add_data(@_);
return $stat->max();} | sub maxdex
{ require Statistics::Descriptive;
my $stat = Statistics::Descriptive::Full->new();
$stat->add_data(@_);
return $stat->maxdex();} | sub mean
{ require Statistics::Descriptive;
my $stat = Statistics::Descriptive::Full->new();
$stat->add_data(@_);
return $stat->mean();} | sub median
{ require Statistics::Descriptive;
my $stat = Statistics::Descriptive::Full->new();
$stat->add_data(@_);
return $stat->median();} | sub min
{ require Statistics::Descriptive;
my $stat = Statistics::Descriptive::Full->new();
$stat->add_data(@_);
return $stat->min();} | sub mindex
{ require Statistics::Descriptive;
my $stat = Statistics::Descriptive::Full->new();
$stat->add_data(@_);
return $stat->mindex();} | sub standard_deviation
{ require Statistics::Descriptive;
my $stat = Statistics::Descriptive::Full->new();
$stat->add_data(@_);
return $stat->standard_deviation();} | sub sum
{ require Statistics::Descriptive;
my $stat = Statistics::Descriptive::Full->new();
$stat->add_data(@_);
return $stat->sum();} | sub ttest
{ opt_default(paired=>0);
my @args = opt_get(@_);
my $paired = opt_val("paired");
my ($t_value, $p_value, $df);
if($paired){
require Statistics::DependantTTest;
require Statistics::Distributions;
my $stat = new Statistics::DependantTTest;
$stat->load_data('x',@{$args[0]});
$stat->load_data('y',@{$args[1]});
($t_value,$df) = $stat->perform_t_test('x','y');
$p_value = Statistics::Distributions::tprob ($df, $t_value);
}else{
require Statistics::TTest;
my $stat = new Statistics::TTest;
$stat->load_data(@args);
$t_value = $stat->t_statistic;
$p_value = $stat->{t_prob};
$df = $stat->df;
}
if(wantarray()){
return ($t_value, $p_value, $df);
}else{
return ($p_value);
}} | sub variance
{ require Statistics::Descriptive;
my $stat = Statistics::Descriptive::Full->new();
$stat->add_data(@_);
return $stat->variance();} | General documentation
Kazuharu Arakawa, <gaou@sfc.keio.ac.jp>
Copyright 2007 by Kazuharu Arakawa
This library is a part of G-language GAE.