line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
package Statistics::Descriptive::Weighted; |
2
|
|
|
|
|
|
|
$VERSION = '0.7'; |
3
|
1
|
|
|
1
|
|
42703
|
use Statistics::Descriptive; |
|
1
|
|
|
|
|
22300
|
|
|
1
|
|
|
|
|
42
|
|
4
|
1
|
|
|
1
|
|
1327
|
use Data::Dumper; |
|
1
|
|
|
|
|
21359
|
|
|
1
|
|
|
|
|
513
|
|
5
|
|
|
|
|
|
|
|
6
|
|
|
|
|
|
|
package Statistics::Descriptive::Weighted::Sparse; |
7
|
1
|
|
|
1
|
|
10
|
use strict; |
|
1
|
|
|
|
|
10
|
|
|
1
|
|
|
|
|
47
|
|
8
|
1
|
|
|
1
|
|
5
|
use vars qw($AUTOLOAD @ISA %fields); |
|
1
|
|
|
|
|
2
|
|
|
1
|
|
|
|
|
85
|
|
9
|
|
|
|
|
|
|
|
10
|
|
|
|
|
|
|
@ISA = qw(Statistics::Descriptive::Sparse); |
11
|
|
|
|
|
|
|
|
12
|
1
|
|
|
1
|
|
5
|
use Carp qw(cluck confess); |
|
1
|
|
|
|
|
1
|
|
|
1
|
|
|
|
|
987
|
|
13
|
|
|
|
|
|
|
|
14
|
|
|
|
|
|
|
##Define a new field to be used as method, to |
15
|
|
|
|
|
|
|
##augment the ones inherited |
16
|
|
|
|
|
|
|
%fields = ( |
17
|
|
|
|
|
|
|
mean => 0, |
18
|
|
|
|
|
|
|
weight => 0, |
19
|
|
|
|
|
|
|
sum_squares => 0, |
20
|
|
|
|
|
|
|
weight_homozyg => 0, |
21
|
|
|
|
|
|
|
biased_variance => 0, |
22
|
|
|
|
|
|
|
biased_standard_deviation => 0 |
23
|
|
|
|
|
|
|
); |
24
|
|
|
|
|
|
|
|
25
|
|
|
|
|
|
|
__PACKAGE__->_make_accessors( [ grep { $_ ne "weight" } keys(%fields) ] ); |
26
|
|
|
|
|
|
|
__PACKAGE__->_make_private_accessors(["weight"]); |
27
|
|
|
|
|
|
|
|
28
|
|
|
|
|
|
|
##Have to override the base method to add new fields to the object |
29
|
|
|
|
|
|
|
##The proxy method from base class is still valid |
30
|
|
|
|
|
|
|
sub new { |
31
|
15
|
|
|
15
|
|
3322
|
my $proto = shift; |
32
|
15
|
|
33
|
|
|
87
|
my $class = ref($proto) || $proto; |
33
|
15
|
|
|
|
|
87
|
my $self = $class->SUPER::new(); ##Create my self re SUPER |
34
|
15
|
|
|
|
|
500
|
@{ $self->{'_permitted'} } {keys %fields} = values %fields; |
|
15
|
|
|
|
|
65
|
|
35
|
15
|
|
|
|
|
52
|
@{ $self } {keys %fields} = values %fields; |
|
15
|
|
|
|
|
43
|
|
36
|
15
|
|
|
|
|
47
|
bless ($self, $class); #Re-anneal the object |
37
|
15
|
|
|
|
|
55
|
return $self; |
38
|
|
|
|
|
|
|
} |
39
|
|
|
|
|
|
|
|
40
|
|
|
|
|
|
|
sub add_data { |
41
|
22
|
|
|
22
|
|
43
|
my $self = shift; ##Myself |
42
|
22
|
|
|
|
|
35
|
my $oldmean; |
43
|
|
|
|
|
|
|
my $oldweight; |
44
|
0
|
|
|
|
|
0
|
my ($min,$max); |
45
|
0
|
|
|
|
|
0
|
my $aref; |
46
|
|
|
|
|
|
|
|
47
|
22
|
100
|
66
|
|
|
182
|
if ( (not ref $_[0] eq 'ARRAY') || (exists $_[1] and (not (ref $_[1] eq 'ARRAY') || @{$_[0]} != @{$_[1]} ) ) ) { |
|
|
|
66
|
|
|
|
|
|
|
|
66
|
|
|
|
|
48
|
4
|
|
|
|
|
599
|
cluck "WARNING: Expected input are two references to two arrays of equal length; first data, then positive weights. Second array is optional.\n"; |
49
|
4
|
|
|
|
|
468
|
return undef; |
50
|
|
|
|
|
|
|
} |
51
|
|
|
|
|
|
|
|
52
|
18
|
|
|
|
|
36
|
my ($datum,$weight) = @_; |
53
|
|
|
|
|
|
|
|
54
|
|
|
|
|
|
|
##Calculate new mean, pseudo-variance, min and max; |
55
|
|
|
|
|
|
|
##The on-line weighted incremental algorithm for variance is based on West 1979 from Wikipedia |
56
|
|
|
|
|
|
|
##D. H. D. West (1979). Communications of the ACM, 22, 9, 532-535: Updating Mean and Variance Estimates: An Improved Method |
57
|
|
|
|
|
|
|
|
58
|
|
|
|
|
|
|
## NEW in Version 0.4: |
59
|
|
|
|
|
|
|
## I calculate a sample weighted variance based on normalized weights rather than the sample size |
60
|
|
|
|
|
|
|
## correction factor is: 1 / (1 - sum [w_i / (sum w_i) ]^2) |
61
|
|
|
|
|
|
|
|
62
|
|
|
|
|
|
|
## call H = sum [w_i / (sum w_i) ]^2. An online update eq for H is H_new = (sum.w_old^2 * H_old) + weight^2) / sum.w^2 |
63
|
|
|
|
|
|
|
|
64
|
|
|
|
|
|
|
## correction factor is then 1 / (1 - H_new) |
65
|
|
|
|
|
|
|
|
66
|
18
|
|
|
|
|
22
|
my $weighterror; |
67
|
18
|
|
|
|
|
55
|
for (0..$#$datum ) { |
68
|
247
|
100
|
|
|
|
539
|
if (not defined $$weight[$_]) { |
69
|
4
|
|
|
|
|
8
|
$$weight[$_] = 1; |
70
|
|
|
|
|
|
|
} |
71
|
247
|
100
|
|
|
|
530
|
if ($$weight[$_] <= 0) { |
72
|
1
|
|
|
|
|
10
|
$weighterror = 1; |
73
|
1
|
|
|
|
|
3
|
next; |
74
|
|
|
|
|
|
|
} |
75
|
246
|
|
|
|
|
355
|
$oldmean = $self->{mean}; |
76
|
246
|
|
|
|
|
285
|
$oldweight = $self->{weight}; |
77
|
246
|
|
|
|
|
343
|
$self->{weight} += $$weight[$_]; |
78
|
246
|
|
|
|
|
653
|
$self->{weight_homozyg} = ((($oldweight ** 2 * $self->{weight_homozyg}) + $$weight[$_] ** 2) / ( $self->{weight} ** 2 )); |
79
|
246
|
|
|
|
|
295
|
$self->{count}++; |
80
|
246
|
|
|
|
|
400
|
$self->{sum} += ($$weight[$_] * $$datum[$_]); |
81
|
246
|
|
|
|
|
628
|
$self->{mean} += (($$weight[$_] / $self->{weight} ) * ($$datum[$_] - $oldmean)); |
82
|
246
|
|
|
|
|
600
|
$self->{sum_squares} += (($$weight[$_] / $self->{weight} ) * ($$datum[$_] - $oldmean) ** 2) * $oldweight; |
83
|
246
|
100
|
100
|
|
|
1108
|
if (not defined $self->{max} or $$datum[$_] > $self->{max}) { |
84
|
243
|
|
|
|
|
368
|
$self->{max} = $$datum[$_]; |
85
|
|
|
|
|
|
|
} |
86
|
246
|
100
|
100
|
|
|
1042
|
if (not defined $self->{min} or $$datum[$_] < $self->{min}) { |
87
|
15
|
|
|
|
|
32
|
$self->{min} = $$datum[$_]; |
88
|
|
|
|
|
|
|
} |
89
|
|
|
|
|
|
|
} |
90
|
18
|
100
|
|
|
|
198
|
cluck "WARNING: One or more data with nonpositive weights were skipped.\n" if ($weighterror); |
91
|
18
|
|
|
|
|
179
|
$self->{sample_range} = $self->{max} - $self->{min}; |
92
|
|
|
|
|
|
|
|
93
|
18
|
100
|
|
|
|
48
|
if ($self->{count} > 1) { |
94
|
14
|
|
|
|
|
41
|
$self->{variance} = ($self->{sum_squares} / ((1 - $self->{weight_homozyg}) * $self->{weight})); |
95
|
14
|
|
|
|
|
43
|
$self->{standard_deviation} = sqrt( $self->{variance}); |
96
|
14
|
|
|
|
|
30
|
$self->{biased_variance} = ($self->{sum_squares} / $self->{weight}); |
97
|
14
|
|
|
|
|
28
|
$self->{biased_standard_deviation} = sqrt( $self->{biased_variance}); |
98
|
|
|
|
|
|
|
} |
99
|
18
|
|
|
|
|
67
|
return 1; |
100
|
|
|
|
|
|
|
} |
101
|
|
|
|
|
|
|
|
102
|
|
|
|
|
|
|
sub weight { |
103
|
12
|
|
|
12
|
|
8374
|
my $self = shift; |
104
|
12
|
100
|
|
|
|
37
|
if (@_ > 0) { |
105
|
2
|
|
|
|
|
399
|
cluck "WARNING: Sparse statistics object expects zero arguments to weight function, returns sum of weights."; |
106
|
|
|
|
|
|
|
} |
107
|
12
|
|
|
|
|
228
|
return $self->_weight(); |
108
|
|
|
|
|
|
|
} |
109
|
|
|
|
|
|
|
|
110
|
|
|
|
|
|
|
## OVERRIDES FOR UNSUPPORTED FUNCTIONS |
111
|
|
|
|
|
|
|
|
112
|
|
|
|
|
|
|
sub mindex{ |
113
|
0
|
|
|
0
|
|
0
|
confess "ERROR: Statistics::Descriptive::Weighted does not support this function."; |
114
|
|
|
|
|
|
|
} |
115
|
|
|
|
|
|
|
|
116
|
|
|
|
|
|
|
sub maxdex{ |
117
|
0
|
|
|
0
|
|
0
|
confess "ERROR: Statistics::Descriptive::Weighted does not support this function."; |
118
|
|
|
|
|
|
|
} |
119
|
|
|
|
|
|
|
|
120
|
|
|
|
|
|
|
1; |
121
|
|
|
|
|
|
|
|
122
|
|
|
|
|
|
|
package Statistics::Descriptive::Weighted::Full; |
123
|
|
|
|
|
|
|
|
124
|
1
|
|
|
1
|
|
6
|
use Carp qw(cluck confess); |
|
1
|
|
|
|
|
3
|
|
|
1
|
|
|
|
|
62
|
|
125
|
1
|
|
|
1
|
|
1218
|
use Tree::Treap; |
|
1
|
|
|
|
|
3
|
|
|
1
|
|
|
|
|
36
|
|
126
|
1
|
|
|
1
|
|
10
|
use strict; |
|
1
|
|
|
|
|
1
|
|
|
1
|
|
|
|
|
45
|
|
127
|
1
|
|
|
1
|
|
5
|
use vars qw(@ISA %fields); |
|
1
|
|
|
|
|
3
|
|
|
1
|
|
|
|
|
3590
|
|
128
|
|
|
|
|
|
|
|
129
|
|
|
|
|
|
|
@ISA = qw(Statistics::Descriptive::Weighted::Sparse); |
130
|
|
|
|
|
|
|
|
131
|
|
|
|
|
|
|
##Create a list of fields not to remove when data is updated |
132
|
|
|
|
|
|
|
%fields = ( |
133
|
|
|
|
|
|
|
_permitted => undef, ##Place holder for the inherited key hash |
134
|
|
|
|
|
|
|
data => undef, ##keys from variate values to a hashref with keys weight, cdf, tail-prob |
135
|
|
|
|
|
|
|
did_cdf => undef, ##flag to indicate whether CDF/quantile fun has been computed or not |
136
|
|
|
|
|
|
|
quantile => undef, ##"hash" for quantile function |
137
|
|
|
|
|
|
|
percentile => undef, ##"hash" for percentile function |
138
|
|
|
|
|
|
|
maxweight => 0, |
139
|
|
|
|
|
|
|
mode => undef, |
140
|
|
|
|
|
|
|
order => 1, |
141
|
|
|
|
|
|
|
_reserved => undef, ##Place holder for this lookup hash |
142
|
|
|
|
|
|
|
); |
143
|
|
|
|
|
|
|
|
144
|
|
|
|
|
|
|
##Have to override the base method to add the data to the object |
145
|
|
|
|
|
|
|
##The proxy method from above is still valid |
146
|
|
|
|
|
|
|
sub new { |
147
|
8
|
|
|
8
|
|
3455
|
my $proto = shift; |
148
|
8
|
|
33
|
|
|
43
|
my $class = ref($proto) || $proto; |
149
|
8
|
|
|
|
|
39
|
my $self = $class->SUPER::new(); ##Create my self re SUPER |
150
|
8
|
|
|
|
|
41
|
$self->{data} = new Tree::Treap("num"); ## inserts data by numeric comparison |
151
|
8
|
|
|
|
|
16
|
$self->{did_cdf} = 0; |
152
|
8
|
|
|
|
|
25
|
$self->{maxweight} = 0; |
153
|
8
|
|
|
|
|
27
|
$self->{quantile} = new Tree::Treap("num"); |
154
|
8
|
|
|
|
|
31
|
$self->{percentile} = new Tree::Treap("num"); |
155
|
8
|
|
|
|
|
18
|
$self->{order} = 1; |
156
|
8
|
|
|
|
|
17
|
$self->{'_reserved'} = \%fields; |
157
|
8
|
|
|
|
|
21
|
bless ($self, $class); |
158
|
8
|
|
|
|
|
41
|
return $self; |
159
|
|
|
|
|
|
|
} |
160
|
|
|
|
|
|
|
|
161
|
|
|
|
|
|
|
## The treap gives relatively fast search and good performance on possibly sorted data |
162
|
|
|
|
|
|
|
## The choice is motivated by heavy intended use for Empirical Distribution Function |
163
|
|
|
|
|
|
|
## A lot of work is done at insertion for faster computation on search |
164
|
|
|
|
|
|
|
## THE ACTUAL DATA INSERTION IS DONE AT FUNCTION _addweight |
165
|
|
|
|
|
|
|
## The data structure loses information. Like a hash keys appear only once. |
166
|
|
|
|
|
|
|
## The value of a key is its sum of weight for that key, and the cumulative weight |
167
|
|
|
|
|
|
|
sub add_data { |
168
|
9
|
|
|
9
|
|
20
|
my $self = shift; |
169
|
9
|
|
|
|
|
11
|
my $key; |
170
|
|
|
|
|
|
|
|
171
|
9
|
50
|
33
|
|
|
95
|
if ( (not ref $_[0] eq 'ARRAY') || (exists $_[1] and (not (ref $_[1] eq 'ARRAY') || @{$_[0]} != @{$_[1]} ) ) ) { |
|
|
|
66
|
|
|
|
|
|
|
|
33
|
|
|
|
|
172
|
0
|
|
|
|
|
0
|
cluck "WARNING: Expected input are two references to two arrays of equal length; first data, then positive weights. Second array is optional.\n"; |
173
|
0
|
|
|
|
|
0
|
return undef; |
174
|
|
|
|
|
|
|
} |
175
|
|
|
|
|
|
|
|
176
|
9
|
|
|
|
|
152
|
my ($datum,$weight) = @_; |
177
|
9
|
|
|
|
|
14
|
my $filterdatum = []; |
178
|
9
|
|
|
|
|
26
|
my $filterweight = []; |
179
|
9
|
|
|
|
|
12
|
my $weighterror; |
180
|
|
|
|
|
|
|
my $newweight; |
181
|
9
|
|
|
|
|
30
|
for (0..$#$datum) { |
182
|
222
|
100
|
|
|
|
517
|
if (not defined $$weight[$_]) { |
183
|
5
|
|
|
|
|
14
|
$$weight[$_] = 1; |
184
|
|
|
|
|
|
|
} |
185
|
222
|
100
|
|
|
|
619
|
if ($$weight[$_] > 0) { |
186
|
221
|
|
|
|
|
422
|
push @$filterdatum,$$datum[$_]; |
187
|
221
|
|
|
|
|
1240
|
push @$filterweight,$$weight[$_]; |
188
|
221
|
|
|
|
|
558
|
$newweight = $self->_addweight($$datum[$_], $$weight[$_]); |
189
|
221
|
100
|
|
|
|
614
|
if ($newweight > $self->{maxweight}) { |
190
|
9
|
|
|
|
|
14
|
$self->{maxweight} = $newweight; |
191
|
9
|
|
|
|
|
61
|
$self->{mode} = $$datum[$_]; |
192
|
|
|
|
|
|
|
} |
193
|
|
|
|
|
|
|
} |
194
|
|
|
|
|
|
|
else { |
195
|
1
|
|
|
|
|
4
|
$weighterror = 1; |
196
|
|
|
|
|
|
|
} |
197
|
|
|
|
|
|
|
} |
198
|
9
|
100
|
|
|
|
264
|
cluck "WARNING: One or more data with nonpositive weights were skipped.\n" if ($weighterror); |
199
|
9
|
|
|
|
|
265
|
$self->SUPER::add_data($filterdatum,$filterweight); ##Perform base statistics on the data |
200
|
|
|
|
|
|
|
##Clear the did_cdf flag |
201
|
9
|
|
|
|
|
15
|
$self->{did_cdf} = 0; |
202
|
|
|
|
|
|
|
##Need to delete all cached keys |
203
|
9
|
|
|
|
|
16
|
foreach $key (keys %{ $self }) { # Check each key in the object |
|
9
|
|
|
|
|
78
|
|
204
|
|
|
|
|
|
|
# If it's a reserved key for this class, keep it |
205
|
223
|
100
|
|
|
|
460
|
next if exists $self->{'_reserved'}->{$key}; |
206
|
|
|
|
|
|
|
# If it comes from the base class, keep it |
207
|
143
|
100
|
|
|
|
342
|
next if exists $self->{'_permitted'}->{$key}; |
208
|
8
|
|
|
|
|
21
|
delete $self->{$key}; # Delete the out of date cached key |
209
|
|
|
|
|
|
|
} |
210
|
9
|
|
|
|
|
131
|
return 1; |
211
|
|
|
|
|
|
|
} |
212
|
|
|
|
|
|
|
|
213
|
|
|
|
|
|
|
sub count { |
214
|
232
|
|
|
232
|
|
2223
|
my $self = shift; |
215
|
232
|
100
|
|
|
|
448
|
if (@_ == 1) { ##Inquire |
|
|
50
|
|
|
|
|
|
216
|
221
|
|
|
|
|
699
|
my $val = $self->{data}->get_val($_[0]); |
217
|
221
|
100
|
|
|
|
1099
|
return (defined $val ? ${ $val }{'count'} : $val); |
|
1
|
|
|
|
|
5
|
|
218
|
|
|
|
|
|
|
} |
219
|
|
|
|
|
|
|
elsif (@_ == 0) { ##Inquire |
220
|
11
|
|
|
|
|
39
|
return $self->{count}; |
221
|
|
|
|
|
|
|
} |
222
|
|
|
|
|
|
|
else { |
223
|
0
|
|
|
|
|
0
|
cluck "WARNING: Only 1 or fewer arguments expected."; |
224
|
|
|
|
|
|
|
} |
225
|
0
|
|
|
|
|
0
|
return 1; |
226
|
|
|
|
|
|
|
} |
227
|
|
|
|
|
|
|
|
228
|
|
|
|
|
|
|
sub weight { |
229
|
442
|
|
|
442
|
|
1704
|
my $self = shift; |
230
|
442
|
100
|
|
|
|
901
|
if (@_ == 1) { ##Inquire |
|
|
50
|
|
|
|
|
|
231
|
440
|
|
|
|
|
1438
|
my $val = $self->{data}->get_val($_[0]); |
232
|
440
|
100
|
|
|
|
1462
|
return (defined $val ? ${ $val }{'weight'} : $val); |
|
219
|
|
|
|
|
599
|
|
233
|
|
|
|
|
|
|
} |
234
|
|
|
|
|
|
|
elsif (@_ == 0) { ##Inquire |
235
|
2
|
|
|
|
|
11
|
return $self->{weight}; |
236
|
|
|
|
|
|
|
} |
237
|
|
|
|
|
|
|
else { |
238
|
0
|
|
|
|
|
0
|
cluck "WARNING: Only 1 or fewer arguments expected."; |
239
|
|
|
|
|
|
|
} |
240
|
0
|
|
|
|
|
0
|
return 1; |
241
|
|
|
|
|
|
|
} |
242
|
|
|
|
|
|
|
|
243
|
|
|
|
|
|
|
sub _addweight { |
244
|
221
|
|
|
221
|
|
251
|
my $self = shift; |
245
|
221
|
|
100
|
|
|
513
|
my $oldweight = ($self->weight($_[0]) || 0); |
246
|
221
|
|
|
|
|
310
|
my $newweight = $_[1] + $oldweight; |
247
|
221
|
|
|
|
|
761
|
my $value = $self->{data}->get_val($_[0]); |
248
|
221
|
100
|
|
|
|
541
|
my $weights = ($value ? $$value{'weights'} : [] ); |
249
|
221
|
|
|
|
|
491
|
push @$weights, $_[1]; |
250
|
221
|
100
|
|
|
|
469
|
my $orders = ($value ? $$value{'order'} : [] ); |
251
|
221
|
|
|
|
|
461
|
push @$orders, $self->{order}++; |
252
|
221
|
|
100
|
|
|
473
|
my $newcount = ($self->count($_[0]) || 0) + 1; |
253
|
221
|
50
|
|
|
|
411
|
if (@_ == 2) { ##Assign |
254
|
221
|
|
|
|
|
1196
|
my $values = {'weight' => $newweight, 'weights' => $weights, 'count' => $newcount, 'order' => $orders, 'cdf' => undef, 'rt_tail_prob' => undef, 'percentile' => undef}; |
255
|
221
|
|
|
|
|
812
|
$self->{data}->insert($_[0],$values); |
256
|
|
|
|
|
|
|
} |
257
|
|
|
|
|
|
|
else { |
258
|
0
|
|
|
|
|
0
|
cluck "WARNING: Only two arguments (key, addend) expected."; |
259
|
|
|
|
|
|
|
} |
260
|
221
|
|
|
|
|
448
|
return $newweight; |
261
|
|
|
|
|
|
|
} |
262
|
|
|
|
|
|
|
|
263
|
|
|
|
|
|
|
sub _do_cdf { |
264
|
6
|
|
|
6
|
|
10
|
my $self = shift; |
265
|
6
|
|
|
|
|
10
|
my $cumweight = 0; |
266
|
6
|
|
|
|
|
29
|
foreach my $key ($self->{data}->keys()){ |
267
|
217
|
|
|
|
|
650
|
my $value = $self->{data}->get_val($key); |
268
|
217
|
|
|
|
|
447
|
my $keyweight = $self->weight($key); |
269
|
|
|
|
|
|
|
|
270
|
217
|
|
|
|
|
272
|
my $oldcumweight = $cumweight; |
271
|
217
|
|
|
|
|
262
|
$cumweight += $keyweight; |
272
|
|
|
|
|
|
|
|
273
|
217
|
|
|
|
|
375
|
my $propcumweight = $cumweight / $self->{weight}; |
274
|
217
|
|
|
|
|
359
|
my $right_tail_prob = (1 - ($oldcumweight / $self->{weight})); |
275
|
217
|
|
|
|
|
350
|
my $percentile = ((100 / $self->{weight}) * ($cumweight - ($keyweight / 2))); |
276
|
|
|
|
|
|
|
|
277
|
217
|
|
|
|
|
342
|
$$value{'cdf'} = $propcumweight; |
278
|
217
|
|
|
|
|
303
|
$$value{'rt_tail_prob'} = $right_tail_prob; |
279
|
217
|
|
|
|
|
4108
|
$$value{'percentile'} = $percentile; |
280
|
|
|
|
|
|
|
|
281
|
217
|
|
|
|
|
825
|
$self->{data}->insert($key,$value); |
282
|
217
|
|
|
|
|
667
|
$self->{quantile}->insert($propcumweight,$key); |
283
|
217
|
|
|
|
|
633
|
$self->{percentile}->insert($percentile,$key); |
284
|
|
|
|
|
|
|
} |
285
|
6
|
|
|
|
|
120
|
$self->{did_cdf} = 1; |
286
|
6
|
|
|
|
|
13
|
return 1; |
287
|
|
|
|
|
|
|
} |
288
|
|
|
|
|
|
|
|
289
|
|
|
|
|
|
|
sub quantile { |
290
|
11
|
|
|
11
|
|
295
|
my $self = shift; |
291
|
11
|
100
|
|
|
|
36
|
$self->_do_cdf() unless $self->{did_cdf}; |
292
|
11
|
50
|
|
|
|
27
|
if (@_ == 1) { ##Inquire |
293
|
11
|
|
|
|
|
13
|
my $proportion = shift; |
294
|
11
|
50
|
33
|
|
|
55
|
cluck "WARNING: Expects an argument between 0 and 1 inclusive." if ($proportion < 0 or $proportion > 1); |
295
|
11
|
|
|
|
|
40
|
my @keys = $self->{quantile}->range_keys($proportion, undef); |
296
|
11
|
|
|
|
|
15
|
my $key = $keys[0]; ## GET THE SMALLEST QUANTILE g.e. $proportion |
297
|
11
|
|
|
|
|
33
|
return $self->{quantile}->get_val($keys[0]); |
298
|
|
|
|
|
|
|
} |
299
|
|
|
|
|
|
|
else { |
300
|
0
|
|
|
|
|
0
|
cluck "WARNING: Exactly 1 argument expected."; |
301
|
0
|
|
|
|
|
0
|
return undef; |
302
|
|
|
|
|
|
|
} |
303
|
|
|
|
|
|
|
} |
304
|
|
|
|
|
|
|
|
305
|
|
|
|
|
|
|
sub percentile { |
306
|
4
|
|
|
4
|
|
7
|
my $self = shift; |
307
|
4
|
100
|
|
|
|
16
|
$self->_do_cdf() unless $self->{did_cdf}; |
308
|
4
|
50
|
|
|
|
12
|
if (@_ != 1) { |
309
|
0
|
|
|
|
|
0
|
cluck "WARNING: Exactly 1 argument expected."; |
310
|
|
|
|
|
|
|
} |
311
|
4
|
|
|
|
|
5
|
my $percent = shift; |
312
|
4
|
50
|
33
|
|
|
20
|
if ($percent < 0 or $percent > 100) { |
313
|
0
|
|
|
|
|
0
|
cluck "WARNING: Expects an argument between 0 and 100 inclusive."; |
314
|
|
|
|
|
|
|
} |
315
|
4
|
|
|
|
|
6
|
my $percentile; |
316
|
4
|
50
|
|
|
|
19
|
if ($percent < $self->{percentile}->minimum()) { |
|
|
50
|
|
|
|
|
|
317
|
0
|
|
|
|
|
0
|
$percentile = $self->{data}->minimum(); |
318
|
|
|
|
|
|
|
} elsif ($percent > $self->{percentile}->maximum()) { |
319
|
0
|
|
|
|
|
0
|
$percentile = $self->{data}->maximum(); |
320
|
|
|
|
|
|
|
} else { |
321
|
4
|
|
|
|
|
15
|
my @lekeys = $self->{percentile}->range_keys(undef,$percent); |
322
|
4
|
|
|
|
|
8
|
my $lekey = $lekeys[-1]; |
323
|
4
|
|
|
|
|
18
|
my @gekeys = $self->{percentile}->range_keys($percent, undef); |
324
|
4
|
|
|
|
|
7
|
my $gekey = $gekeys[0]; |
325
|
4
|
|
|
|
|
16
|
my $leval = $self->{percentile}->get_val($lekey); |
326
|
4
|
|
|
|
|
6
|
$percentile = $leval; |
327
|
4
|
100
|
|
|
|
13
|
if ($gekey != $lekey) { |
328
|
3
|
|
|
|
|
11
|
my $geval = $self->{percentile}->get_val($gekey); |
329
|
3
|
|
|
|
|
15
|
$percentile += ($percent - $lekey) / ($gekey - $lekey) * ($geval - $leval); |
330
|
|
|
|
|
|
|
} |
331
|
|
|
|
|
|
|
} |
332
|
4
|
|
|
|
|
22
|
return $percentile; |
333
|
|
|
|
|
|
|
} |
334
|
|
|
|
|
|
|
|
335
|
|
|
|
|
|
|
|
336
|
|
|
|
|
|
|
sub median { |
337
|
2
|
|
|
2
|
|
4
|
my $self = shift; |
338
|
|
|
|
|
|
|
|
339
|
|
|
|
|
|
|
##Cached? |
340
|
2
|
50
|
|
|
|
19
|
return $self->{median} if defined $self->{median}; |
341
|
2
|
|
|
|
|
7
|
return $self->{median} = $self->percentile(50); |
342
|
|
|
|
|
|
|
} |
343
|
|
|
|
|
|
|
|
344
|
|
|
|
|
|
|
sub mode { |
345
|
1
|
|
|
1
|
|
306
|
my $self = shift; |
346
|
1
|
|
|
|
|
6
|
return $self->{mode}; |
347
|
|
|
|
|
|
|
} |
348
|
|
|
|
|
|
|
|
349
|
|
|
|
|
|
|
sub cdf { |
350
|
11
|
|
|
11
|
|
56
|
my $self = shift; |
351
|
11
|
50
|
|
|
|
34
|
$self->_do_cdf() unless $self->{did_cdf}; |
352
|
11
|
50
|
|
|
|
29
|
if (@_ == 1) { ##Inquire |
353
|
11
|
|
|
|
|
16
|
my $value = shift; |
354
|
11
|
100
|
|
|
|
41
|
return 0 if ($self->{data}->minimum() > $value); |
355
|
10
|
|
|
|
|
38
|
my @keys = $self->{data}->range_keys(undef, $value); |
356
|
10
|
|
|
|
|
37
|
my $key = $keys[-1]; ## GET THE LARGEST OBSERVED VALUE l.e. $value |
357
|
10
|
|
|
|
|
17
|
return ${ $self->{data}->get_val($key) }{'cdf'}; |
|
10
|
|
|
|
|
45
|
|
358
|
|
|
|
|
|
|
} |
359
|
|
|
|
|
|
|
else { |
360
|
0
|
|
|
|
|
0
|
cluck "WARNING: Exactly 1 argument expected."; |
361
|
0
|
|
|
|
|
0
|
return undef; |
362
|
|
|
|
|
|
|
} |
363
|
|
|
|
|
|
|
} |
364
|
|
|
|
|
|
|
|
365
|
|
|
|
|
|
|
sub survival { |
366
|
4
|
|
|
4
|
|
7
|
my $self = shift; |
367
|
4
|
50
|
|
|
|
12
|
$self->_do_cdf() unless $self->{did_cdf}; |
368
|
4
|
50
|
|
|
|
10
|
if (@_ == 1) { ##Inquire |
369
|
4
|
|
|
|
|
6
|
my $value = shift; |
370
|
4
|
100
|
|
|
|
13
|
return 1 if ($self->{data}->minimum() > $value); |
371
|
3
|
|
|
|
|
12
|
my @keys = $self->{data}->range_keys(undef, $value); |
372
|
3
|
|
|
|
|
5
|
my $key = $keys[-1]; ## GET THE LARGEST OBSERVED VALUE l.e. $value |
373
|
3
|
|
|
|
|
4
|
return 1 - (${ $self->{data}->get_val($key) }{'cdf'}); |
|
3
|
|
|
|
|
9
|
|
374
|
|
|
|
|
|
|
} |
375
|
|
|
|
|
|
|
else { |
376
|
0
|
|
|
|
|
0
|
cluck "WARNING: Only 1 argument expected."; |
377
|
0
|
|
|
|
|
0
|
return undef; |
378
|
|
|
|
|
|
|
} |
379
|
|
|
|
|
|
|
} |
380
|
|
|
|
|
|
|
|
381
|
|
|
|
|
|
|
sub rtp { |
382
|
6
|
|
|
6
|
|
9
|
my $self = shift; |
383
|
6
|
50
|
|
|
|
18
|
$self->_do_cdf() unless $self->{did_cdf}; |
384
|
6
|
50
|
|
|
|
17
|
if (@_ == 1) { ##Inquire |
385
|
6
|
|
|
|
|
7
|
my $value = shift; |
386
|
6
|
100
|
|
|
|
22
|
return 0 if ($self->{data}->maximum() < $value); |
387
|
3
|
|
|
|
|
13
|
my @keys = $self->{data}->range_keys($value, undef); |
388
|
3
|
|
|
|
|
6
|
my $key = $keys[0]; ## GET THE SMALLEST OBSERVED VALUE g.e. $value |
389
|
3
|
|
|
|
|
5
|
return ${ $self->{data}->get_val($key) }{'rt_tail_prob'}; |
|
3
|
|
|
|
|
9
|
|
390
|
|
|
|
|
|
|
} |
391
|
|
|
|
|
|
|
else { |
392
|
0
|
|
|
|
|
0
|
cluck "WARNING: Only 1 argument expected."; |
393
|
0
|
|
|
|
|
0
|
return undef; |
394
|
|
|
|
|
|
|
} |
395
|
|
|
|
|
|
|
} |
396
|
|
|
|
|
|
|
|
397
|
|
|
|
|
|
|
sub get_data { |
398
|
1
|
|
|
1
|
|
2
|
my $self = shift; |
399
|
1
|
50
|
|
|
|
7
|
$self->_do_cdf() unless $self->{did_cdf}; |
400
|
1
|
|
|
|
|
6
|
my ($uniqkeys, $sumweights, $keys, $weights, $counts, $cdfs, $rtps, $percentiles, $order) = ([],[],[],[],[],[],[],[],[]); |
401
|
1
|
|
|
|
|
6
|
my $key = $self->{'data'}->minimum(); |
402
|
1
|
|
|
|
|
5
|
while (defined $key){ |
403
|
5
|
|
|
|
|
15
|
my $value = $self->{data}->get_val($key); |
404
|
5
|
|
|
|
|
10
|
push @$uniqkeys, $key; |
405
|
5
|
|
|
|
|
15
|
push @$sumweights, $$value{'weight'}; |
406
|
5
|
|
|
|
|
32
|
foreach my $weight (@{ $$value{'weights'} } ) { |
|
5
|
|
|
|
|
12
|
|
407
|
6
|
|
|
|
|
11
|
push @$keys, $key; |
408
|
6
|
|
|
|
|
12
|
push @$weights, $weight; |
409
|
|
|
|
|
|
|
} |
410
|
5
|
|
|
|
|
8
|
push @$order, @{ $$value{'order'}}; |
|
5
|
|
|
|
|
10
|
|
411
|
5
|
|
|
|
|
9
|
push @$counts, $$value{'count'}; |
412
|
5
|
|
|
|
|
8
|
push @$cdfs, $$value{'cdf'}; |
413
|
5
|
|
|
|
|
6
|
push @$rtps, $$value{'rt_tail_prob'}; |
414
|
5
|
|
|
|
|
22
|
push @$percentiles, $$value{'percentile'}; |
415
|
5
|
|
|
|
|
16
|
$key = $self->{'data'}->successor($key); |
416
|
|
|
|
|
|
|
} |
417
|
1
|
|
|
|
|
26
|
return {'uniqvars' => $uniqkeys, 'sumweights' => $sumweights, 'counts' => $counts, 'cdfs' => $cdfs, 'rtps' => $rtps, 'vars' => $keys, 'weights' => $weights, 'percentiles' => $percentiles, 'order' => $order}; |
418
|
|
|
|
|
|
|
} |
419
|
|
|
|
|
|
|
|
420
|
|
|
|
|
|
|
sub print { |
421
|
1
|
|
|
1
|
|
294
|
my $self = shift; |
422
|
1
|
|
|
|
|
5
|
print Data::Dumper->Dump([$self->get_data()]); |
423
|
|
|
|
|
|
|
} |
424
|
|
|
|
|
|
|
|
425
|
|
|
|
|
|
|
## OVERRIDES FOR UNSUPPORTED FUNCTIONS |
426
|
|
|
|
|
|
|
|
427
|
|
|
|
|
|
|
sub sort_data{ |
428
|
0
|
|
|
0
|
|
|
confess "ERROR: Statistics::Descriptive::Weighted does not support this function."; |
429
|
|
|
|
|
|
|
} |
430
|
|
|
|
|
|
|
|
431
|
|
|
|
|
|
|
sub presorted{ |
432
|
0
|
|
|
0
|
|
|
confess "ERROR: Statistics::Descriptive::Weighted does not support this function."; |
433
|
|
|
|
|
|
|
} |
434
|
|
|
|
|
|
|
|
435
|
|
|
|
|
|
|
sub harmonic_mean{ |
436
|
0
|
|
|
0
|
|
|
confess "ERROR: Statistics::Descriptive::Weighted does not support this function."; |
437
|
|
|
|
|
|
|
} |
438
|
|
|
|
|
|
|
|
439
|
|
|
|
|
|
|
sub geometric_mean{ |
440
|
0
|
|
|
0
|
|
|
confess "ERROR: Statistics::Descriptive::Weighted does not support this function."; |
441
|
|
|
|
|
|
|
} |
442
|
|
|
|
|
|
|
|
443
|
|
|
|
|
|
|
sub trimmed_mean{ |
444
|
0
|
|
|
0
|
|
|
confess "ERROR: Statistics::Descriptive::Weighted does not support this function."; |
445
|
|
|
|
|
|
|
} |
446
|
|
|
|
|
|
|
|
447
|
|
|
|
|
|
|
sub frequency_distribution{ |
448
|
0
|
|
|
0
|
|
|
confess "ERROR: Statistics::Descriptive::Weighted does not support this function."; |
449
|
|
|
|
|
|
|
} |
450
|
|
|
|
|
|
|
|
451
|
|
|
|
|
|
|
sub least_squares_fit{ |
452
|
0
|
|
|
0
|
|
|
confess "ERROR: Statistics::Descriptive::Weighted does not support this function."; |
453
|
|
|
|
|
|
|
} |
454
|
|
|
|
|
|
|
|
455
|
|
|
|
|
|
|
|
456
|
|
|
|
|
|
|
1; |
457
|
|
|
|
|
|
|
|
458
|
|
|
|
|
|
|
package Statistics::Descriptive; |
459
|
|
|
|
|
|
|
|
460
|
|
|
|
|
|
|
##All modules return true. |
461
|
|
|
|
|
|
|
1; |
462
|
|
|
|
|
|
|
|
463
|
|
|
|
|
|
|
__END__ |