forked from npshub/mantid
-
Notifications
You must be signed in to change notification settings - Fork 0
/
StatisticsTest.h
311 lines (262 loc) · 9.53 KB
/
StatisticsTest.h
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
// Mantid Repository : https://github.com/mantidproject/mantid
//
// Copyright © 2018 ISIS Rutherford Appleton Laboratory UKRI,
// NScD Oak Ridge National Laboratory, European Spallation Source,
// Institut Laue - Langevin & CSNS, Institute of High Energy Physics, CAS
// SPDX - License - Identifier: GPL - 3.0 +
#pragma once
#include "MantidKernel/Statistics.h"
#include <algorithm>
#include <cmath>
#include <cxxtest/TestSuite.h>
#include <string>
#include <vector>
using namespace Mantid::Kernel;
using std::string;
using std::vector;
class StatisticsTest : public CxxTest::TestSuite {
public:
void test_Doubles_And_Default_Flags_Calculates_All_Stats() {
vector<double> data;
data.emplace_back(17.2);
data.emplace_back(18.1);
data.emplace_back(16.5);
data.emplace_back(18.3);
data.emplace_back(12.6);
Statistics stats = getStatistics(data);
TS_ASSERT_EQUALS(stats.mean, 16.54);
TS_ASSERT_DELTA(stats.standard_deviation, 2.0732, 0.0001);
TS_ASSERT_EQUALS(stats.minimum, 12.6);
TS_ASSERT_EQUALS(stats.maximum, 18.3);
TS_ASSERT_EQUALS(stats.median, 17.2);
}
void test_Doubles_With_Sorted_Data() {
vector<double> data;
data.emplace_back(17.2);
data.emplace_back(18.1);
data.emplace_back(16.5);
data.emplace_back(18.3);
data.emplace_back(12.6);
sort(data.begin(), data.end());
Statistics stats = getStatistics(data, (StatOptions::Median | StatOptions::SortedData));
TS_ASSERT(std::isnan(stats.mean));
TS_ASSERT(std::isnan(stats.standard_deviation));
TS_ASSERT(std::isnan(stats.minimum));
TS_ASSERT(std::isnan(stats.maximum));
TS_ASSERT_EQUALS(stats.median, 17.2);
}
void test_Unsorted_Data_With_Sorted_Flag_Gives_Expected_Result_For_Median() {
vector<double> data;
data.emplace_back(17.2); // Median value
data.emplace_back(18.1);
data.emplace_back(16.5);
data.emplace_back(18.3);
data.emplace_back(12.6);
Statistics stats = getStatistics(data, (StatOptions::Median | StatOptions::SortedData));
TS_ASSERT(std::isnan(stats.mean));
TS_ASSERT(std::isnan(stats.standard_deviation));
TS_ASSERT(std::isnan(stats.minimum));
TS_ASSERT(std::isnan(stats.maximum));
TS_ASSERT_EQUALS(stats.median, 17.2);
}
void test_Doubles_With_Corrected_StdDev_Calculates_Mean() {
vector<double> data;
data.emplace_back(17.2);
data.emplace_back(18.1);
data.emplace_back(16.5);
data.emplace_back(18.3);
data.emplace_back(12.6);
sort(data.begin(), data.end());
Statistics stats = getStatistics(data, StatOptions::CorrectedStdDev);
TS_ASSERT_EQUALS(stats.mean, 16.54);
TS_ASSERT_DELTA(stats.standard_deviation, 2.3179, 0.0001);
TS_ASSERT_EQUALS(stats.minimum, 12.6);
TS_ASSERT_EQUALS(stats.maximum, 18.3);
TS_ASSERT(std::isnan(stats.median));
}
void test_Types_Can_Be_Disabled_With_Flags() {
vector<double> data;
data.emplace_back(17.2);
data.emplace_back(18.1);
data.emplace_back(16.5);
data.emplace_back(18.3);
data.emplace_back(12.6);
Statistics justMean = getStatistics(data, StatOptions::Mean);
TS_ASSERT_EQUALS(justMean.mean, 16.54);
TS_ASSERT(std::isnan(justMean.standard_deviation));
TS_ASSERT(std::isnan(justMean.minimum));
TS_ASSERT(std::isnan(justMean.maximum));
TS_ASSERT(std::isnan(justMean.median));
}
void testZscores() {
vector<double> data;
data.emplace_back(12);
data.emplace_back(13);
data.emplace_back(9);
data.emplace_back(18);
data.emplace_back(7);
data.emplace_back(9);
data.emplace_back(14);
data.emplace_back(16);
data.emplace_back(10);
data.emplace_back(12);
data.emplace_back(7);
data.emplace_back(13);
data.emplace_back(14);
data.emplace_back(19);
data.emplace_back(10);
data.emplace_back(16);
data.emplace_back(12);
data.emplace_back(16);
data.emplace_back(19);
data.emplace_back(11);
std::vector<double> Zscore = getZscore(data);
TS_ASSERT_DELTA(Zscore[4], 1.6397, 0.0001);
TS_ASSERT_DELTA(Zscore[6], 0.3223, 0.0001);
std::vector<double> ZModscore = getModifiedZscore(data);
TS_ASSERT_DELTA(ZModscore[4], 1.2365, 0.0001);
TS_ASSERT_DELTA(ZModscore[6], 0.3372, 0.0001);
}
void testDoubleSingle() {
vector<double> data;
data.emplace_back(42.);
Statistics stats = getStatistics(data);
TS_ASSERT_EQUALS(stats.mean, 42.);
TS_ASSERT_EQUALS(stats.standard_deviation, 0.);
TS_ASSERT_EQUALS(stats.minimum, 42.);
TS_ASSERT_EQUALS(stats.maximum, 42.);
TS_ASSERT_EQUALS(stats.median, 42.);
}
void testInt32Even() {
vector<int32_t> data;
data.emplace_back(1);
data.emplace_back(2);
data.emplace_back(3);
data.emplace_back(4);
data.emplace_back(5);
data.emplace_back(6);
Statistics stats = getStatistics(data);
TS_ASSERT_EQUALS(stats.mean, 3.5);
TS_ASSERT_DELTA(stats.standard_deviation, 1.7078, 0.0001);
TS_ASSERT_EQUALS(stats.minimum, 1.);
TS_ASSERT_EQUALS(stats.maximum, 6.);
TS_ASSERT_EQUALS(stats.median, 3.5);
}
bool my_isnan(const double number) { return number != number; }
void testString() {
vector<string> data{"hi there"};
Statistics stats = getStatistics(data);
TS_ASSERT(my_isnan(stats.mean));
TS_ASSERT(my_isnan(stats.standard_deviation));
TS_ASSERT(my_isnan(stats.minimum));
TS_ASSERT(my_isnan(stats.maximum));
TS_ASSERT(my_isnan(stats.median));
}
/** Test function to calculate Rwp
*/
void testRwp() {
vector<double> obsY(4);
vector<double> calY(4);
vector<double> obsE(4);
obsY[0] = 1.0;
calY[0] = 1.1;
obsE[0] = 1.0;
obsY[1] = 2.0;
calY[1] = 2.1;
obsE[1] = 1.2;
obsY[2] = 3.0;
calY[2] = 3.5;
obsE[2] = 1.4;
obsY[3] = 1.0;
calY[3] = 1.3;
obsE[3] = 1.0;
Rfactor rfactor = getRFactor(obsY, calY, obsE);
TS_ASSERT_DELTA(rfactor.Rwp, 0.1582, 0.0001);
}
/** Test throw exception
*/
void testRwpException1() {
vector<double> obsY{1.0, 2.0, 3.0, 1.0};
vector<double> calY{1.1, 2.1, 3.5, 1.3};
vector<double> obsE{1.0, 1.2, 1.4};
TS_ASSERT_THROWS_ANYTHING(getRFactor(obsY, calY, obsE));
}
/** Test throw exception on empty array
*/
void testRwpException2() {
vector<double> obsY;
vector<double> calY;
vector<double> obsE;
TS_ASSERT_THROWS_ANYTHING(getRFactor(obsY, calY, obsE));
}
/// Test moment calculations about origin and mean
void test_getMoments() {
const double mean = 5.;
const double sigma = 4.;
const double deltaX = .2;
const size_t numX = 200;
// calculate to have same number of points left and right of function
const double offsetX = mean - (.5 * deltaX * static_cast<double>(numX));
// variance about origin
double expVar = mean * mean + sigma * sigma;
// skew about origin
double expSkew = mean * mean * mean + 3. * mean * sigma * sigma;
// x-values to try out
vector<double> x;
for (size_t i = 0; i < numX; ++i)
x.emplace_back(static_cast<double>(i) * deltaX + offsetX);
// just declare so we can have test of exception handling
vector<double> y;
TS_ASSERT_THROWS(getMomentsAboutOrigin(x, y), const std::out_of_range &);
// now calculate the y-values
for (size_t i = 0; i < numX; ++i) {
double temp = (x[i] - mean) / sigma;
y.emplace_back(exp(-.5 * temp * temp) / (sigma * sqrt(2. * M_PI)));
}
// Normal distribution values are taken from the wikipedia page
{
std::cout << "Normal distribution about origin\n";
vector<double> aboutOrigin = getMomentsAboutOrigin(x, y);
TS_ASSERT_EQUALS(aboutOrigin.size(), 4);
TS_ASSERT_DELTA(aboutOrigin[0], 1., .0001);
TS_ASSERT_DELTA(aboutOrigin[1], mean, .0001);
TS_ASSERT_DELTA(aboutOrigin[2], expVar, .001 * expVar);
TS_ASSERT_DELTA(aboutOrigin[3], expSkew, .001 * expSkew);
std::cout << "Normal distribution about mean\n";
vector<double> aboutMean = getMomentsAboutMean(x, y);
TS_ASSERT_EQUALS(aboutMean.size(), 4);
TS_ASSERT_DELTA(aboutMean[0], 1., .0001);
TS_ASSERT_DELTA(aboutMean[1], 0., .0001);
TS_ASSERT_DELTA(aboutMean[2], sigma * sigma, .001 * expVar);
TS_ASSERT_DELTA(aboutMean[3], 0., .0001 * expSkew);
}
// Now a gaussian function as a histogram
y.clear();
for (size_t i = 0; i < numX - 1; ++i) // one less y than x makes it a histogram
{
double templeft = (x[i] - mean) / sigma;
templeft = exp(-.5 * templeft * templeft) / (sigma * sqrt(2. * M_PI));
double tempright = (x[i + 1] - mean) / sigma;
tempright = exp(-.5 * tempright * tempright) / (sigma * sqrt(2. * M_PI));
y.emplace_back(.5 * deltaX * (templeft + tempright));
// std::cout << i << ":\t" << x[i] << "\t" << y[i] << '\n';
}
// Normal distribution values are taken from the wikipedia page
{
std::cout << "Normal distribution about origin\n";
vector<double> aboutOrigin = getMomentsAboutOrigin(x, y);
TS_ASSERT_EQUALS(aboutOrigin.size(), 4);
TS_ASSERT_DELTA(aboutOrigin[0], 1., .0001);
TS_ASSERT_DELTA(aboutOrigin[1], mean, .0001);
TS_ASSERT_DELTA(aboutOrigin[2], expVar, .001 * expVar);
TS_ASSERT_DELTA(aboutOrigin[3], expSkew, .001 * expSkew);
std::cout << "Normal distribution about mean\n";
vector<double> aboutMean = getMomentsAboutMean(x, y);
TS_ASSERT_EQUALS(aboutMean.size(), 4);
TS_ASSERT_DELTA(aboutMean[0], 1., .0001);
TS_ASSERT_DELTA(aboutMean[1], 0., .0001);
TS_ASSERT_DELTA(aboutMean[2], sigma * sigma, .001 * expVar);
TS_ASSERT_DELTA(aboutMean[3], 0., .0001 * expSkew);
}
}
};