diff options
Diffstat (limited to 'test/de/lmu/ifi/dbs/elki/math/TestStatisticalMoments.java')
| -rw-r--r-- | test/de/lmu/ifi/dbs/elki/math/TestStatisticalMoments.java | 120 |
1 files changed, 0 insertions, 120 deletions
diff --git a/test/de/lmu/ifi/dbs/elki/math/TestStatisticalMoments.java b/test/de/lmu/ifi/dbs/elki/math/TestStatisticalMoments.java deleted file mode 100644 index fe63b707..00000000 --- a/test/de/lmu/ifi/dbs/elki/math/TestStatisticalMoments.java +++ /dev/null @@ -1,120 +0,0 @@ -package de.lmu.ifi.dbs.elki.math; - -/* - This file is part of ELKI: - Environment for Developing KDD-Applications Supported by Index-Structures - - Copyright (C) 2012 - Ludwig-Maximilians-Universität München - Lehr- und Forschungseinheit für Datenbanksysteme - ELKI Development Team - - This program is free software: you can redistribute it and/or modify - it under the terms of the GNU Affero General Public License as published by - the Free Software Foundation, either version 3 of the License, or - (at your option) any later version. - - This program is distributed in the hope that it will be useful, - but WITHOUT ANY WARRANTY; without even the implied warranty of - MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - GNU Affero General Public License for more details. - - You should have received a copy of the GNU Affero General Public License - along with this program. If not, see <http://www.gnu.org/licenses/>. - */ - -import static org.junit.Assert.assertEquals; - -import org.junit.Test; - -import de.lmu.ifi.dbs.elki.JUnit4Test; - -/** - * Unit test to check our statistical moments class for correctness. - * - * @author Erich Schubert - */ -public class TestStatisticalMoments implements JUnit4Test { - @Test - public void testTiny() { - double[] data = new double[] { 222, 1122, 45444 }; - StatisticalMoments m = new StatisticalMoments(); - for (double d : data) { - m.put(d); - } - // Validated with R package "e1071" and scipy. - // Looks as we might be closer to scipys result! - // skewness(c(222,1122,45444), type = 1) = 0.70614285367114104 - // scipy.stats.skew([222, 1122, 45444], bias=True) = 0.7061428536711412 - assertEquals("Naive skewness does not match.", 0.70614285367114104, m.getNaiveSkewness(), 1e-15); - // skewness(c(222,1122,45444), type = 2) = 1.7296896770071024 - // scipy.stats.skew([222, 1122, 45444], bias=False) = 1.7296896770071026 - assertEquals("Sample skewness does not match.", 1.7296896770071024, m.getSampleSkewness(), 1e-15); - // Type 3 skewness is not in ELKI. - - // kurtosis(c(222,1122,45444), type = 1) = -1.5 - assertEquals("Naive kurtosis does not match.", -1.5, m.getNaiveExcessKurtosis(), 1e-15); - assertEquals("Naive kurtosis does not match.", -1.5 + 3.0, m.getNaiveKurtosis(), 1e-15); - // type = 2 needs 4 observations - so do we! - } - - @Test - public void testGamma() { - double[] data = new double[] { // 100 Gamma(5,5) distributed random values. - 0.6388530607016163, 0.9782442336324599, 0.8882176966052615, 0.7849646645745156, 2.0633275427120497, // - 0.9545195282320902, 1.092390620469477, 0.7258265676258062, 1.378026049091643, 0.8310918468931533,// - 0.9495339766440013, 0.7126363476416752, 0.5608700366395489, 0.31408795273142415, 1.4909322345449254, // - 0.46625313007571856, 1.1870054487101156, 0.7640843133327799, 2.235582041645692, 1.5812487824715005, // - 1.512895069952923, 0.606517358771931, 1.1961880366469766, 0.7854465499137, 0.8084781536051177, // - 0.7615405607968533, 1.6614435713861262, 1.4869842072109933, 0.43465864880894045, 2.007341464773332, // - 0.33478515273215137, 1.9093706181977907, 0.3539177300096471, 0.44888559667524086, 1.0249621186169293, // - 0.38784905783431306, 0.7632290555534638, 0.9941038353388855, 0.6358745449186729, 0.5750658437507069, // - 0.9565742499910938, 1.2221328658541637, 0.9389920544737386, 1.21461175894598, 1.5713316877580166, // - 0.3470088915002977, 0.4419291224056868, 2.343465732285217, 1.0612049261527454, 0.9960347794653911, // - 0.5433216826433932, 1.006345176861023, 1.034059277904943, 0.7199916430119295, 1.218878192200847, // - 1.0023101563986212, 1.093784991451296, 0.9748868952634213, 1.81732410812918, 0.5259404003069172, // - 1.5179842038671814, 0.7461497981977685, 0.47530253061596, 0.810558795121139, 0.9383514512691828, // - 0.42371144806092004, 0.8698616597894346, 0.4541860955288436, 0.3136304450065802, 0.8799371863836344, // - 0.9854734113277808, 0.8651302802119106, 0.7981146204118504, 0.5255964404563404, 0.8570668629471919, // - 1.0294303406520284, 1.4592029352371467, 1.1034285928922425, 1.0042662401564635, 0.6443339212869743, // - 1.4096313764251664, 0.9328112988095885, 2.0874718257265483, 1.063702918067901, 0.5885221900074733, // - 1.216856018361022, 0.5572986786142953, 0.6620626132872955, 0.7357210048560157, 1.5174141129183254, // - 1.2293542801457167, 1.490026433254134, 1.4307462549666328, 1.381520418446136, 0.48400651328811123, // - 0.8448303085270435, 1.1514319532981445, 0.8944580168473978, 0.5003052782124537, 1.1457096934062219 }; - StatisticalMoments m = new StatisticalMoments(); - for (double d : data) { - m.put(d); - } - // Validated with R package "e1071" and scipy: - // Looks as we might be closer to scipys result! - // skewness(data, type = 1) = 0.8553563631726689 - // scipy.stats.skew(data, bias=True) = 0.8553563631726712 - assertEquals("Naive skewness does not match.", 0.8553563631726689, m.getNaiveSkewness(), 1e-14); - // skewness(data, type = 2) = 0.868437587353074 - // scipy.stats.skew(data, bias=False) = 0.868437587353077 - assertEquals("Sample skewness does not match.", 0.868437587353074, m.getSampleSkewness(), 1e-14); - // Type 3 skewness is not in ELKI. - - // kurtosis(data, type = 1) = 0.4778717586207377 - // scipy.stats.kurtosis(data, fisher=True) = 0.4778717586207377 - assertEquals("Naive kurtosis does not match.", 0.4778717586207377, m.getNaiveExcessKurtosis(), 1e-14); - assertEquals("Naive kurtosis does not match.", 0.4778717586207377 + 3.0, m.getNaiveKurtosis(), 1e-14); - // kurtosis(data, type = 2) = 0.565141985530060675 - // scipy.stats.kurtosis(data, fisher=True, bias=False) = 0.5651419855300603 - // scipy.stats.kurtosis(data, fisher=True, bias=False) = 0.5651419855300603 - // scipy.stats.kurtosis(data, fisher=False, bias=False) = 3.5651419855300603 - assertEquals("Sample kurtosis does not match.", 0.565141985530060675, m.getSampleExcessKurtosis(), 1e-14); - assertEquals("Sample kurtosis does not match.", 3.565141985530060675, m.getSampleKurtosis(), 1e-14); - - // numpy.array(data).mean() = 0.98342960290360215 - assertEquals("Mean does not match.", 0.98342960290360215, m.getMean(), 1e-15); - // numpy.array(data).std() = 0.44953213330440578 - assertEquals("Naive stddev does not match.", 0.44953213330440578, m.getNaiveStddev(), 1e-15); - // numpy.array(data).std(ddof=1) = 0.45179679314507293 - assertEquals("Sample stddev does not match.", 0.45179679314507293, m.getSampleStddev(), 1e-15); - // numpy.array(data).min() = 0.31363044500658022 - assertEquals("Min does not match.", 0.31363044500658022, m.getMin(), 1e-16); - // numpy.array(data).max() = 2.343465732285217 - assertEquals("Max does not match.", 2.343465732285217, m.getMax(), 1e-16); - } -} |