1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
|
#!/usr/local/bin/perl
# ********************************************************************
# * Copyright (C) 2016 and later: Unicode, Inc. and others.
# * License & terms of use: http://www.unicode.org/copyright.html
# ********************************************************************
# ********************************************************************
# * COPYRIGHT:
# * Copyright (c) 2002, International Business Machines Corporation and
# * others. All Rights Reserved.
# ********************************************************************
package Dataset;
use Statistics::Descriptive;
use Statistics::Distributions;
use strict;
# Create a new Dataset with the given data.
sub new {
my ($class) = shift;
my $self = bless {
_data => \@_,
_scale => 1.0,
_mean => 0.0,
_error => 0.0,
}, $class;
my $n = @_;
if ($n >= 1) {
my $stats = Statistics::Descriptive::Full->new();
$stats->add_data(@{$self->{_data}});
$self->{_mean} = $stats->mean();
if ($n >= 2) {
# Use a t distribution rather than Gaussian because (a) we
# assume an underlying normal dist, (b) we do not know the
# standard deviation -- we estimate it from the data, and (c)
# we MAY have a small sample size (also works for large n).
my $t = Statistics::Distributions::tdistr($n-1, 0.005);
$self->{_error} = $t * $stats->standard_deviation();
}
}
$self;
}
# Set a scaling factor for all data; 1.0 means no scaling.
# Scale must be > 0.
sub setScale {
my ($self, $scale) = @_;
$self->{_scale} = $scale;
}
# Multiply the scaling factor by a value.
sub scaleBy {
my ($self, $a) = @_;
$self->{_scale} *= $a;
}
# Return the mean.
sub getMean {
my $self = shift;
return $self->{_mean} * $self->{_scale};
}
# Return a 99% error based on the t distribution. The dataset
# is described as getMean() +/- getError().
sub getError {
my $self = shift;
return $self->{_error} * $self->{_scale};
}
# Divide two Datasets and return a new one, maintaining the
# mean+/-error. The new Dataset has no data points.
sub divide {
my $self = shift;
my $rhs = shift;
my $minratio = ($self->{_mean} - $self->{_error}) /
($rhs->{_mean} + $rhs->{_error});
my $maxratio = ($self->{_mean} + $self->{_error}) /
($rhs->{_mean} - $rhs->{_error});
my $result = Dataset->new();
$result->{_mean} = ($minratio + $maxratio) / 2;
$result->{_error} = $result->{_mean} - $minratio;
$result->{_scale} = $self->{_scale} / $rhs->{_scale};
$result;
}
# subtracts two Datasets and return a new one, maintaining the
# mean+/-error. The new Dataset has no data points.
sub subtract {
my $self = shift;
my $rhs = shift;
my $result = Dataset->new();
$result->{_mean} = $self->{_mean} - $rhs->{_mean};
$result->{_error} = $self->{_error} + $rhs->{_error};
$result->{_scale} = $self->{_scale};
$result;
}
# adds two Datasets and return a new one, maintaining the
# mean+/-error. The new Dataset has no data points.
sub add {
my $self = shift;
my $rhs = shift;
my $result = Dataset->new();
$result->{_mean} = $self->{_mean} + $rhs->{_mean};
$result->{_error} = $self->{_error} + $rhs->{_error};
$result->{_scale} = $self->{_scale};
$result;
}
# Divides a dataset by a scalar.
# The new Dataset has no data points.
sub divideByScalar {
my $self = shift;
my $s = shift;
my $result = Dataset->new();
$result->{_mean} = $self->{_mean}/$s;
$result->{_error} = $self->{_error}/$s;
$result->{_scale} = $self->{_scale};
$result;
}
# Divides a dataset by a scalar.
# The new Dataset has no data points.
sub multiplyByScalar {
my $self = shift;
my $s = shift;
my $result = Dataset->new();
$result->{_mean} = $self->{_mean}*$s;
$result->{_error} = $self->{_error}*$s;
$result->{_scale} = $self->{_scale};
$result;
}
1;
|
