1 | /*************************************************************************
|
---|
2 | Cephes Math Library Release 2.8: June, 2000
|
---|
3 | Copyright 1984, 1987, 1995, 2000 by Stephen L. Moshier
|
---|
4 |
|
---|
5 | Contributors:
|
---|
6 | * Sergey Bochkanov (ALGLIB project). Translation from C to
|
---|
7 | pseudocode.
|
---|
8 |
|
---|
9 | See subroutines comments for additional copyrights.
|
---|
10 |
|
---|
11 | >>> SOURCE LICENSE >>>
|
---|
12 | This program is free software; you can redistribute it and/or modify
|
---|
13 | it under the terms of the GNU General Public License as published by
|
---|
14 | the Free Software Foundation (www.fsf.org); either version 2 of the
|
---|
15 | License, or (at your option) any later version.
|
---|
16 |
|
---|
17 | This program is distributed in the hope that it will be useful,
|
---|
18 | but WITHOUT ANY WARRANTY; without even the implied warranty of
|
---|
19 | MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
---|
20 | GNU General Public License for more details.
|
---|
21 |
|
---|
22 | A copy of the GNU General Public License is available at
|
---|
23 | http://www.fsf.org/licensing/licenses
|
---|
24 |
|
---|
25 | >>> END OF LICENSE >>>
|
---|
26 | *************************************************************************/
|
---|
27 |
|
---|
28 | using System;
|
---|
29 |
|
---|
30 | namespace alglib
|
---|
31 | {
|
---|
32 | public class binomialdistr
|
---|
33 | {
|
---|
34 | /*************************************************************************
|
---|
35 | Binomial distribution
|
---|
36 |
|
---|
37 | Returns the sum of the terms 0 through k of the Binomial
|
---|
38 | probability density:
|
---|
39 |
|
---|
40 | k
|
---|
41 | -- ( n ) j n-j
|
---|
42 | > ( ) p (1-p)
|
---|
43 | -- ( j )
|
---|
44 | j=0
|
---|
45 |
|
---|
46 | The terms are not summed directly; instead the incomplete
|
---|
47 | beta integral is employed, according to the formula
|
---|
48 |
|
---|
49 | y = bdtr( k, n, p ) = incbet( n-k, k+1, 1-p ).
|
---|
50 |
|
---|
51 | The arguments must be positive, with p ranging from 0 to 1.
|
---|
52 |
|
---|
53 | ACCURACY:
|
---|
54 |
|
---|
55 | Tested at random points (a,b,p), with p between 0 and 1.
|
---|
56 |
|
---|
57 | a,b Relative error:
|
---|
58 | arithmetic domain # trials peak rms
|
---|
59 | For p between 0.001 and 1:
|
---|
60 | IEEE 0,100 100000 4.3e-15 2.6e-16
|
---|
61 |
|
---|
62 | Cephes Math Library Release 2.8: June, 2000
|
---|
63 | Copyright 1984, 1987, 1995, 2000 by Stephen L. Moshier
|
---|
64 | *************************************************************************/
|
---|
65 | public static double binomialdistribution(int k,
|
---|
66 | int n,
|
---|
67 | double p)
|
---|
68 | {
|
---|
69 | double result = 0;
|
---|
70 | double dk = 0;
|
---|
71 | double dn = 0;
|
---|
72 |
|
---|
73 | System.Diagnostics.Debug.Assert((double)(p)>=(double)(0) & (double)(p)<=(double)(1), "Domain error in BinomialDistribution");
|
---|
74 | System.Diagnostics.Debug.Assert(k>=-1 & k<=n, "Domain error in BinomialDistribution");
|
---|
75 | if( k==-1 )
|
---|
76 | {
|
---|
77 | result = 0;
|
---|
78 | return result;
|
---|
79 | }
|
---|
80 | if( k==n )
|
---|
81 | {
|
---|
82 | result = 1;
|
---|
83 | return result;
|
---|
84 | }
|
---|
85 | dn = n-k;
|
---|
86 | if( k==0 )
|
---|
87 | {
|
---|
88 | dk = Math.Pow(1.0-p, dn);
|
---|
89 | }
|
---|
90 | else
|
---|
91 | {
|
---|
92 | dk = k+1;
|
---|
93 | dk = ibetaf.incompletebeta(dn, dk, 1.0-p);
|
---|
94 | }
|
---|
95 | result = dk;
|
---|
96 | return result;
|
---|
97 | }
|
---|
98 |
|
---|
99 |
|
---|
100 | /*************************************************************************
|
---|
101 | Complemented binomial distribution
|
---|
102 |
|
---|
103 | Returns the sum of the terms k+1 through n of the Binomial
|
---|
104 | probability density:
|
---|
105 |
|
---|
106 | n
|
---|
107 | -- ( n ) j n-j
|
---|
108 | > ( ) p (1-p)
|
---|
109 | -- ( j )
|
---|
110 | j=k+1
|
---|
111 |
|
---|
112 | The terms are not summed directly; instead the incomplete
|
---|
113 | beta integral is employed, according to the formula
|
---|
114 |
|
---|
115 | y = bdtrc( k, n, p ) = incbet( k+1, n-k, p ).
|
---|
116 |
|
---|
117 | The arguments must be positive, with p ranging from 0 to 1.
|
---|
118 |
|
---|
119 | ACCURACY:
|
---|
120 |
|
---|
121 | Tested at random points (a,b,p).
|
---|
122 |
|
---|
123 | a,b Relative error:
|
---|
124 | arithmetic domain # trials peak rms
|
---|
125 | For p between 0.001 and 1:
|
---|
126 | IEEE 0,100 100000 6.7e-15 8.2e-16
|
---|
127 | For p between 0 and .001:
|
---|
128 | IEEE 0,100 100000 1.5e-13 2.7e-15
|
---|
129 |
|
---|
130 | Cephes Math Library Release 2.8: June, 2000
|
---|
131 | Copyright 1984, 1987, 1995, 2000 by Stephen L. Moshier
|
---|
132 | *************************************************************************/
|
---|
133 | public static double binomialcdistribution(int k,
|
---|
134 | int n,
|
---|
135 | double p)
|
---|
136 | {
|
---|
137 | double result = 0;
|
---|
138 | double dk = 0;
|
---|
139 | double dn = 0;
|
---|
140 |
|
---|
141 | System.Diagnostics.Debug.Assert((double)(p)>=(double)(0) & (double)(p)<=(double)(1), "Domain error in BinomialDistributionC");
|
---|
142 | System.Diagnostics.Debug.Assert(k>=-1 & k<=n, "Domain error in BinomialDistributionC");
|
---|
143 | if( k==-1 )
|
---|
144 | {
|
---|
145 | result = 1;
|
---|
146 | return result;
|
---|
147 | }
|
---|
148 | if( k==n )
|
---|
149 | {
|
---|
150 | result = 0;
|
---|
151 | return result;
|
---|
152 | }
|
---|
153 | dn = n-k;
|
---|
154 | if( k==0 )
|
---|
155 | {
|
---|
156 | if( (double)(p)<(double)(0.01) )
|
---|
157 | {
|
---|
158 | dk = -nearunityunit.expm1(dn*nearunityunit.log1p(-p));
|
---|
159 | }
|
---|
160 | else
|
---|
161 | {
|
---|
162 | dk = 1.0-Math.Pow(1.0-p, dn);
|
---|
163 | }
|
---|
164 | }
|
---|
165 | else
|
---|
166 | {
|
---|
167 | dk = k+1;
|
---|
168 | dk = ibetaf.incompletebeta(dk, dn, p);
|
---|
169 | }
|
---|
170 | result = dk;
|
---|
171 | return result;
|
---|
172 | }
|
---|
173 |
|
---|
174 |
|
---|
175 | /*************************************************************************
|
---|
176 | Inverse binomial distribution
|
---|
177 |
|
---|
178 | Finds the event probability p such that the sum of the
|
---|
179 | terms 0 through k of the Binomial probability density
|
---|
180 | is equal to the given cumulative probability y.
|
---|
181 |
|
---|
182 | This is accomplished using the inverse beta integral
|
---|
183 | function and the relation
|
---|
184 |
|
---|
185 | 1 - p = incbi( n-k, k+1, y ).
|
---|
186 |
|
---|
187 | ACCURACY:
|
---|
188 |
|
---|
189 | Tested at random points (a,b,p).
|
---|
190 |
|
---|
191 | a,b Relative error:
|
---|
192 | arithmetic domain # trials peak rms
|
---|
193 | For p between 0.001 and 1:
|
---|
194 | IEEE 0,100 100000 2.3e-14 6.4e-16
|
---|
195 | IEEE 0,10000 100000 6.6e-12 1.2e-13
|
---|
196 | For p between 10^-6 and 0.001:
|
---|
197 | IEEE 0,100 100000 2.0e-12 1.3e-14
|
---|
198 | IEEE 0,10000 100000 1.5e-12 3.2e-14
|
---|
199 |
|
---|
200 | Cephes Math Library Release 2.8: June, 2000
|
---|
201 | Copyright 1984, 1987, 1995, 2000 by Stephen L. Moshier
|
---|
202 | *************************************************************************/
|
---|
203 | public static double invbinomialdistribution(int k,
|
---|
204 | int n,
|
---|
205 | double y)
|
---|
206 | {
|
---|
207 | double result = 0;
|
---|
208 | double dk = 0;
|
---|
209 | double dn = 0;
|
---|
210 | double p = 0;
|
---|
211 |
|
---|
212 | System.Diagnostics.Debug.Assert(k>=0 & k<n, "Domain error in InvBinomialDistribution");
|
---|
213 | dn = n-k;
|
---|
214 | if( k==0 )
|
---|
215 | {
|
---|
216 | if( (double)(y)>(double)(0.8) )
|
---|
217 | {
|
---|
218 | p = -nearunityunit.expm1(nearunityunit.log1p(y-1.0)/dn);
|
---|
219 | }
|
---|
220 | else
|
---|
221 | {
|
---|
222 | p = 1.0-Math.Pow(y, 1.0/dn);
|
---|
223 | }
|
---|
224 | }
|
---|
225 | else
|
---|
226 | {
|
---|
227 | dk = k+1;
|
---|
228 | p = ibetaf.incompletebeta(dn, dk, 0.5);
|
---|
229 | if( (double)(p)>(double)(0.5) )
|
---|
230 | {
|
---|
231 | p = ibetaf.invincompletebeta(dk, dn, 1.0-y);
|
---|
232 | }
|
---|
233 | else
|
---|
234 | {
|
---|
235 | p = 1.0-ibetaf.invincompletebeta(dn, dk, y);
|
---|
236 | }
|
---|
237 | }
|
---|
238 | result = p;
|
---|
239 | return result;
|
---|
240 | }
|
---|
241 | }
|
---|
242 | }
|
---|