search.cpan.org is shutting down
Shlomi Fish > Statistics-Descriptive-3.0607 > Statistics::Descriptive::Smoother

Statistics-Descriptive-3.0607.tar.gz

Dependencies

Annotate this POD

Website

# CPAN RT

 New 1 Open 1
View/Report Bugs
Module Version: 3.0607   Source   Latest Release: Statistics-Descriptive-3.0613

# NAME

Statistics::Descriptive::Smoother - Base module for smoothing statistical data

# SYNOPSIS

```  use Statistics::Descriptive::Smoother;
my \$smoother = Statistics::Descriptive::Smoother->instantiate({
method   => 'exponential',
coeff    => 0.5,
data     => [1, 2, 3, 4, 5],
samples  => [110, 120, 130, 140, 150],
});
my @smoothed_data = \$smoother->get_smoothed_data();```

# DESCRIPTION

This module provide methods to smooth the trend of a series of statistical data.

The methods provided are the `Exponential` and the `Weighted Exponential` (see respectively `Statistics::Descriptive::Smoother::Exponential` and `Statistics::Descriptive::Smoother::Weightedexponential` for more details).

This class is just a factory that will instantiate the object to perform the chosen smoothing algorithm.

# METHODS

Statistics::Descriptive::Smoother->instantiate({});

Create a new Smoother object.

This method require several parameters:

method

Method used for the smoothing. Allowed values are: `exponential` and `weightedexponential`

coeff

Smoothing coefficient. It needs to be in the [0;1] range, otherwise undef will be reutrned. `0` means that the series is not smoothed at all, while `1` the series is universally equal to the initial unsmoothed value.

data

Array ref with the data of the series. At least 2 values are needed to smooth the series, undef is returned otherwise.

samples

Array ref with the samples each data value has been built with. This is an optional parameter since it is not used by all the smoothing algorithm.

\$smoother->get_smoothing_coeff();

Returns the smoothing coefficient.

\$smoother->set_smoothing_coeff(0.5);

Set the smoothing coefficient value. It needs to be in the [0;1] range, otherwise undef will be reutrned.

Fabio Ponciroli