NAME¶
Statistics::Test::RandomWalk - Random Walk test for random numbers
SYNOPSIS¶
use Statistics::Test::RandomWalk;
my $tester = Statistics::Test::RandomWalk->new();
$tester->set_data( [map {rand()} 1..1000000] );
my $no_bins = 10;
my ($quant, $got, $expected) = $tester->test($no_bins);
print $tester->data_to_report($quant, $got, $expected);
DESCRIPTION¶
This module implements a Random Walk test of a random number generator as
outlined in Blobel et al (Refer to the SEE ALSO section).
Basically, it tests that the numbers "[0,1]" generated by a random
number generator are distributed evenly. It divides "[0,1]" into
"n" evenly sized bins and calculates the number of expected and
actual random numbers in the bin. (In fact, this counts the cumulated numbers,
but that works the same.)
METHODS¶
new¶
Creates a new random number tester.
set_rescale_factor¶
The default range of the random numbers [0, 1) can be rescaled by a constant
factor. This method is the setter for that factor.
rescale_factor¶
Returns the current rescaling factor.
set_data¶
Sets the random numbers to operate on. First argument must be either an array
reference to an array of random numbers or a code reference.
If the first argument is a code reference, the second argument must be an
integer "n". The code reference is called "n"-times and
its return values are used as random numbers.
The code reference semantics are particularly useful if you do not want to store
all random numbers in memory at the same time. You can write a subroutine
that, for example, generates and returns batches of 100 random numbers so no
more than 101 of these numbers will be in memory at the same time. Note that
if you return 100 numbers at once and pass in "n=50", you will have
a sequence of 5000 random numbers.
test¶
Runs the Random Walk test on the data that was previously set using
"set_data".
First argument must be the number of bins.
Returns three array references. First is an array of quantiles. If the number of
bins was ten, this (and all other returned arrays) will hold ten items.
Second are the determined numbers of random numbers below the quantiles. Third
are the expected counts.
data_to_report¶
From the data returned by the "test()" method, this method creates a
textual report and returns it as a string.
Do not forget to pass in the data that was returned by "test()" or use
the "test_report()" method directly if you do not use the data
otherwise.
SUBROUTINES¶
n_over_k¶
Computes "n" over "k". Uses Perl's big number support and
returns a Math::BigFloat object.
This sub is memoized.
SEE ALSO¶
Math::BigFloat, Memoize, Params::Util
Random number generators: Math::Random::MT, Math::Random, Math::Random::OO,
Math::TrulyRandom, "/dev/random" where available
Statistics::Test::Sequence
The algorithm was taken from: (German)
Blobel, V., and Lohrmann, E.
Statistische und numerische Methoden der
Datenanalyse. Stuttgart, Leipzig: Teubner, 1998
AUTHOR¶
Steffen Mueller, <smueller@cpan.org>
COPYRIGHT AND LICENSE¶
Copyright (C) 2007-2010 by Steffen Mueller
This library is free software; you can redistribute it and/or modify it under
the same terms as Perl itself, either Perl version 5.6 or, at your option, any
later version of Perl 5 you may have available.