1 //===----------------------------------------------------------------------===//
2 //
3 // The LLVM Compiler Infrastructure
4 //
5 // This file is dual licensed under the MIT and the University of Illinois Open
6 // Source Licenses. See LICENSE.TXT for details.
7 //
8 //===----------------------------------------------------------------------===//
9 //
10 // REQUIRES: long_tests
11
12 // <random>
13
14 // template<class RealType = double>
15 // class chi_squared_distribution
16
17 // template<class _URNG> result_type operator()(_URNG& g, const param_type& parm);
18
19 #include <random>
20 #include <cassert>
21 #include <vector>
22 #include <numeric>
23
24 template <class T>
25 inline
26 T
sqr(T x)27 sqr(T x)
28 {
29 return x * x;
30 }
31
main()32 int main()
33 {
34 {
35 typedef std::chi_squared_distribution<> D;
36 typedef D::param_type P;
37 typedef std::minstd_rand G;
38 G g;
39 D d(0.5);
40 P p(1);
41 const int N = 1000000;
42 std::vector<D::result_type> u;
43 for (int i = 0; i < N; ++i)
44 {
45 D::result_type v = d(g, p);
46 assert(d.min() < v);
47 u.push_back(v);
48 }
49 double mean = std::accumulate(u.begin(), u.end(), 0.0) / u.size();
50 double var = 0;
51 double skew = 0;
52 double kurtosis = 0;
53 for (int i = 0; i < u.size(); ++i)
54 {
55 double d = (u[i] - mean);
56 double d2 = sqr(d);
57 var += d2;
58 skew += d * d2;
59 kurtosis += d2 * d2;
60 }
61 var /= u.size();
62 double dev = std::sqrt(var);
63 skew /= u.size() * dev * var;
64 kurtosis /= u.size() * var * var;
65 kurtosis -= 3;
66 double x_mean = p.n();
67 double x_var = 2 * p.n();
68 double x_skew = std::sqrt(8 / p.n());
69 double x_kurtosis = 12 / p.n();
70 assert(std::abs((mean - x_mean) / x_mean) < 0.01);
71 assert(std::abs((var - x_var) / x_var) < 0.01);
72 assert(std::abs((skew - x_skew) / x_skew) < 0.01);
73 assert(std::abs((kurtosis - x_kurtosis) / x_kurtosis) < 0.01);
74 }
75 {
76 typedef std::chi_squared_distribution<> D;
77 typedef D::param_type P;
78 typedef std::mt19937 G;
79 G g;
80 D d(1);
81 P p(2);
82 const int N = 1000000;
83 std::vector<D::result_type> u;
84 for (int i = 0; i < N; ++i)
85 {
86 D::result_type v = d(g, p);
87 assert(d.min() < v);
88 u.push_back(v);
89 }
90 double mean = std::accumulate(u.begin(), u.end(), 0.0) / u.size();
91 double var = 0;
92 double skew = 0;
93 double kurtosis = 0;
94 for (int i = 0; i < u.size(); ++i)
95 {
96 double d = (u[i] - mean);
97 double d2 = sqr(d);
98 var += d2;
99 skew += d * d2;
100 kurtosis += d2 * d2;
101 }
102 var /= u.size();
103 double dev = std::sqrt(var);
104 skew /= u.size() * dev * var;
105 kurtosis /= u.size() * var * var;
106 kurtosis -= 3;
107 double x_mean = p.n();
108 double x_var = 2 * p.n();
109 double x_skew = std::sqrt(8 / p.n());
110 double x_kurtosis = 12 / p.n();
111 assert(std::abs((mean - x_mean) / x_mean) < 0.01);
112 assert(std::abs((var - x_var) / x_var) < 0.01);
113 assert(std::abs((skew - x_skew) / x_skew) < 0.01);
114 assert(std::abs((kurtosis - x_kurtosis) / x_kurtosis) < 0.01);
115 }
116 {
117 typedef std::chi_squared_distribution<> D;
118 typedef D::param_type P;
119 typedef std::minstd_rand G;
120 G g;
121 D d(2);
122 P p(.5);
123 const int N = 1000000;
124 std::vector<D::result_type> u;
125 for (int i = 0; i < N; ++i)
126 {
127 D::result_type v = d(g, p);
128 assert(d.min() < v);
129 u.push_back(v);
130 }
131 double mean = std::accumulate(u.begin(), u.end(), 0.0) / u.size();
132 double var = 0;
133 double skew = 0;
134 double kurtosis = 0;
135 for (int i = 0; i < u.size(); ++i)
136 {
137 double d = (u[i] - mean);
138 double d2 = sqr(d);
139 var += d2;
140 skew += d * d2;
141 kurtosis += d2 * d2;
142 }
143 var /= u.size();
144 double dev = std::sqrt(var);
145 skew /= u.size() * dev * var;
146 kurtosis /= u.size() * var * var;
147 kurtosis -= 3;
148 double x_mean = p.n();
149 double x_var = 2 * p.n();
150 double x_skew = std::sqrt(8 / p.n());
151 double x_kurtosis = 12 / p.n();
152 assert(std::abs((mean - x_mean) / x_mean) < 0.01);
153 assert(std::abs((var - x_var) / x_var) < 0.01);
154 assert(std::abs((skew - x_skew) / x_skew) < 0.01);
155 assert(std::abs((kurtosis - x_kurtosis) / x_kurtosis) < 0.01);
156 }
157 }
158