Statistics::Basic::LeastSquareFit(3pm)

1Statistics::Basic::LeasUtsSeqruaCroenFtirti(b3u)ted PerlStDaotciusmteinctsa:t:iBoansic::LeastSquareFit(3)
2
3
4

NAME

6       Statistics::Basic::LeastSquareFit - find the least square fit for two
7       lists
8

SYNOPSIS

10       A machine to calculate the Least Square Fit of given vectors x and y.
11
12       The module returns the alpha and beta filling this formula:
13
14           $y = $beta * $x + $alpha
15
16       for a given set of x and y co-ordinate pairs.
17
18       Say you have the set of Cartesian coordinates:
19
20           my @points = ( [1,1], [2,2], [3,3], [4,4] );
21
22       The simplest way to find the LSF is as follows:
23
24           my $lsf = lsf()->set_size(int @points);
25              $lsf->insert(@$_) for @points;
26
27       Or this way:
28
29           my $xv  = vector( map {$_->[0]} @points );
30           my $yv  = vector( map {$_->[1]} @points );
31           my $lsf = lsf($xv, $yv);
32
33       And then either query the values or print them like so:
34
35           print "The LSF for $xv and $yv: $lsf\n";
36           my ($yint, $slope) =
37           my ($alpha, $beta) = $lsf->query;
38
39       LSF is meant for finding a line of best fit.  $beta is the slope of the
40       line and $alpha is the y-offset.  Suppose you want to draw the line.
41       Use these to calculate the "x" for a given "y" or vice versa:
42
43           my $y = $lsf->y_given_x( 7 );
44           my $x = $lsf->x_given_y( 7 );
45
46       (Note that x_given_y() can sometimes produce a divide-by-zero error
47       since it has to divide by the $beta.)
48
49       Create a 20 point "moving" LSF like so:
50
51           use Statistics::Basic qw(:all nofill);
52
53           my $sth = $dbh->prepare("select x,y from points where something");
54           my $len = 20;
55           my $lsf = lsf()->set_size($len);
56
57           $sth->execute or die $dbh->errstr;
58           $sth->bind_columns( my ($x, $y) ) or die $dbh->errstr;
59
60           my $count = $len;
61           while( $sth->fetch ) {
62               $lsf->insert( $x, $y );
63               if( defined( my ($yint, $slope) = $lsf->query ) {
64                   print "LSF: y= $slope*x + $yint\n";
65               }
66
67               # This would also work:
68               # print "$lsf\n" if $lsf->query_filled;
69           }
70

METHODS

72       This list of methods skips the methods inherited from
73       Statistics::Basic::_TwoVectorBase (things like insert(), and
74       ginsert()).
75
76       new()
77           Create a new Statistics::Basic::LeastSquareFit object.  This
78           function takes two arguments -- which can either be arrayrefs or
79           Statistics::Basic::Vector objects.  This function is called when
80           the leastsquarefirt() shortcut-function is called.
81
82       query()
83           LSF is meant for finding a line of best fit.  $beta is the slope of
84           the line and $alpha is the y-offset.
85
86               my ($alpha, $beta) = $lsf->query;
87
88       y_given_x()
89           Automatically calculate the y-value on the line for a given
90           x-value.
91
92               my $y = $lsf->y_given_x( 7 );
93
94       x_given_y()
95           Automatically calculate the x-value on the line for a given
96           y-value.
97
98               my $x = $lsf->x_given_y( 7 );
99
100           x_given_y() can sometimes produce a divide-by-zero error since it
101           has to divide by the $beta.  This might be helpful:
102
103               if( defined( my $x = eval { $lsf->x_given_y(7) } ) ) {
104                   warn "there is no x value for 7";
105
106               } else {
107                   print "x (given y=7): $x\n";
108               }
109
110       query_vector1()
111           Return the Statistics::Basic::Vector for the first vector used in
112           the computation of alpha and beta.
113
114       query_vector2()
115           Return the Statistics::Basic::Vector object for the second vector
116           used in the computation of alpha and beta.
117
118       query_mean1()
119           Returns the Statistics::Basic::Mean object for the first vector
120           used in the computation of alpha and beta.
121
122       query_variance1()
123           Returns the Statistics::Basic::Variance object for the first vector
124           used in the computation of alpha and beta.
125
126       query_covariance()
127           Returns the Statistics::Basic::Covariance object used in the
128           computation of alpha and beta.
129

OVERLOADS

131       This object is overloaded.  It tries to return an appropriate string
132       for the calculation, but raises an error in numeric context.
133
134       In boolean context, this object is always true (even when empty).
135

AUTHOR

137       Paul Miller "<jettero@cpan.org>"
138

COPYRIGHT

140       Copyright 2012 Paul Miller -- Licensed under the LGPL
141

NAME

SYNOPSIS

METHODS

OVERLOADS

AUTHOR

COPYRIGHT

SEE ALSO