1 | // $Id: Fisher.cc 2119 2009-12-12 23:11:43Z peter $ |
---|
2 | |
---|
3 | /* |
---|
4 | Copyright (C) 2004 Jari Häkkinen, Peter Johansson |
---|
5 | Copyright (C) 2005 Peter Johansson |
---|
6 | Copyright (C) 2006, 2007, 2008 Jari Häkkinen, Peter Johansson |
---|
7 | Copyright (C) 2009 Peter Johansson |
---|
8 | |
---|
9 | This file is part of the yat library, http://dev.thep.lu.se/yat |
---|
10 | |
---|
11 | The yat library is free software; you can redistribute it and/or |
---|
12 | modify it under the terms of the GNU General Public License as |
---|
13 | published by the Free Software Foundation; either version 3 of the |
---|
14 | License, or (at your option) any later version. |
---|
15 | |
---|
16 | The yat library is distributed in the hope that it will be useful, |
---|
17 | but WITHOUT ANY WARRANTY; without even the implied warranty of |
---|
18 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
---|
19 | General Public License for more details. |
---|
20 | |
---|
21 | You should have received a copy of the GNU General Public License |
---|
22 | along with yat. If not, see <http://www.gnu.org/licenses/>. |
---|
23 | */ |
---|
24 | |
---|
25 | #include "Fisher.h" |
---|
26 | #include "utility.h" |
---|
27 | |
---|
28 | #include <gsl/gsl_cdf.h> |
---|
29 | #include <gsl/gsl_randist.h> |
---|
30 | |
---|
31 | #include <algorithm> |
---|
32 | |
---|
33 | namespace theplu { |
---|
34 | namespace yat { |
---|
35 | namespace statistics { |
---|
36 | |
---|
37 | |
---|
38 | Fisher::Fisher() |
---|
39 | : a_(0), b_(0), c_(0), d_(0), minimum_size_(10), oddsratio_(1.0) |
---|
40 | { |
---|
41 | } |
---|
42 | |
---|
43 | |
---|
44 | Fisher::~Fisher() |
---|
45 | { |
---|
46 | } |
---|
47 | |
---|
48 | |
---|
49 | bool Fisher::calculate_p_exact(void) const |
---|
50 | { |
---|
51 | return ( a_<minimum_size_ || b_<minimum_size_ || |
---|
52 | c_<minimum_size_ || d_<minimum_size_); |
---|
53 | } |
---|
54 | |
---|
55 | double Fisher::Chi2() const |
---|
56 | { |
---|
57 | double a,b,c,d; |
---|
58 | expected(a,b,c,d); |
---|
59 | return (a-a_)*(a-a_)/a + (b-b_)*(b-b_)/b + |
---|
60 | (c-c_)*(c-c_)/c + (d-d_)*(d-d_)/d; |
---|
61 | } |
---|
62 | |
---|
63 | |
---|
64 | void Fisher::expected(double& a, double& b, double& c, double& d) const |
---|
65 | { |
---|
66 | double N = a_+b_+c_+d_; |
---|
67 | a =((a_+b_)*(a_+c_)) / N; |
---|
68 | b =((a_+b_)*(b_+d_)) / N; |
---|
69 | c =((c_+d_)*(a_+c_)) / N; |
---|
70 | d =((c_+d_)*(b_+d_)) / N; |
---|
71 | } |
---|
72 | |
---|
73 | |
---|
74 | unsigned int& Fisher::minimum_size(void) |
---|
75 | { |
---|
76 | return minimum_size_; |
---|
77 | } |
---|
78 | |
---|
79 | |
---|
80 | const unsigned int& Fisher::minimum_size(void) const |
---|
81 | { |
---|
82 | return minimum_size_; |
---|
83 | } |
---|
84 | |
---|
85 | |
---|
86 | double Fisher::oddsratio(const unsigned int a, |
---|
87 | const unsigned int b, |
---|
88 | const unsigned int c, |
---|
89 | const unsigned int d) |
---|
90 | { |
---|
91 | // If a column sum or a row sum is zero, the table is nonsense |
---|
92 | if ((a==0 || d==0) && (c==0 || b==0)){ |
---|
93 | throw std::runtime_error("runtime_error: Table in Fisher is not valid\n"); |
---|
94 | } |
---|
95 | a_ = a; |
---|
96 | b_ = b; |
---|
97 | c_ = c; |
---|
98 | d_ = d; |
---|
99 | |
---|
100 | oddsratio_=static_cast<double>(a*d)/static_cast<double>(b*c); |
---|
101 | return oddsratio_; |
---|
102 | } |
---|
103 | |
---|
104 | |
---|
105 | double Fisher::p_value() const |
---|
106 | { |
---|
107 | if (calculate_p_exact()) |
---|
108 | return p_value_exact(); |
---|
109 | return p_value_approximative(); |
---|
110 | } |
---|
111 | |
---|
112 | |
---|
113 | double Fisher::p_value_one_sided() const |
---|
114 | { |
---|
115 | if (!calculate_p_exact()) { |
---|
116 | if (oddsratio_<1.0) |
---|
117 | return 1.0-p_value_approximative()/2.0; |
---|
118 | return p_value_approximative()/2.0; |
---|
119 | } |
---|
120 | return statistics::cdf_hypergeometric_P(c_, c_+d_, a_+b_, a_+c_); |
---|
121 | } |
---|
122 | |
---|
123 | |
---|
124 | double Fisher::p_value_approximative() const |
---|
125 | { |
---|
126 | return gsl_cdf_chisq_Q(Chi2(), 1.0); |
---|
127 | } |
---|
128 | |
---|
129 | double Fisher::p_value_exact() const |
---|
130 | { |
---|
131 | double res=0; |
---|
132 | double a, c, tmp; |
---|
133 | expected(a, tmp, c, tmp); |
---|
134 | // sum P(x) over x for which abs(x-E(a))>=abs(a-E(a)) |
---|
135 | |
---|
136 | // counting left tail |
---|
137 | int k = static_cast<int>(a - std::abs(a_-a)); |
---|
138 | if (k>=0) |
---|
139 | res = cdf_hypergeometric_P(k, a_+b_, c_+d_, a_+c_); |
---|
140 | // counting right tail |
---|
141 | k = static_cast<int>(c - std::abs(c_-c)); |
---|
142 | if (k>=0) |
---|
143 | res += cdf_hypergeometric_P(k, c_+d_, a_+b_, a_+c_); |
---|
144 | |
---|
145 | // avoid p>1 due to double counting middle one |
---|
146 | return std::min(1.0, res); |
---|
147 | } |
---|
148 | |
---|
149 | |
---|
150 | }}} // of namespace statistics, yat, and theplu |
---|