Differential D22163 Diff 60857 tests/stats/ttest/TTestTest.cpp

Changeset View

Standalone View

View Options

tests/stats/ttest/TTestTest.cpp

  • This file was added.
1​///***************************************************************************
2​// File : CorrelationTest.cpp
3​// Project : LabPlot
4​// Description : Tests for data correlation
5​// --------------------------------------------------------------------
6​// Copyright : (C) 2019 Devanshu Agarwal (agarwaldevanshu8@gmail.com)
7​// ***************************************************************************/
8​
9​///***************************************************************************
10​// * *
11​// * This program is free software; you can redistribute it and/or modify *
12​// * it under the terms of the GNU General Public License as published by *
13​// * the Free Software Foundation; either version 2 of the License, or *
14​// * (at your option) any later version. *
15​// * *
16​// * This program 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 *
19​// * GNU General Public License for more details. *
20​// * *
21​// * You should have received a copy of the GNU General Public License *
22​// * along with this program; if not, write to the Free Software *
23​// * Foundation, Inc., 51 Franklin Street, Fifth Floor, *
24​// * Boston, MA 02110-1301 USA *
25​// * *
26​// ***************************************************************************/
27​
28​#include "TTestTest.h"
29​#include "backend/hypothesisTest/HypothesisTest.h"
30​
31​#include "backend/core/AbstractColumn.h"
32​#include "backend/core/column/Column.h"
33​
34​void TTestTest::twoSampleIndependent_data() {
35​ QTest::addColumn<QVector<double>>("col1Data");
36​ QTest::addColumn<QVector<double>>("col2Data");
37​ QTest::addColumn<double>("tValue_expected");
38​ QTest::addColumn<double>("pValue_expected");
39​
40​ QVector<double> col1Data = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1};
41​ QVector<double> col2Data = {3, 1, 5, 4, 6, 4, 6, 2, 0, 5, 4, 5, 4, 3, 6, 6, 8, 5, 5, 4, 2, 5, 7, 5};
sgerlachUnsubmitted
Done

add comment: "first sample"

sgerlach: add comment: "first sample"
42​ QTest::newRow("ttest1") << col1Data << col2Data << -1.713 << 0.101;
43​
44​ col1Data = {1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
45​ col2Data = {42, 46, 43, 10, 55, 17, 26, 60, 62, 53, 37, 42, 33, 37, 41, 42, 19, 55, 54, 28, 20, 48, 85, 24, 56, 43, 59, 58, 52, 71, 62, 43, 54, 49, 57, 61, 33, 44, 46, 67, 43, 49, 57, 53};
46​ QTest::newRow("ttest2") << col1Data << col2Data << -2.26 << 0.028;
47​}
48​
49​void TTestTest::twoSampleIndependent() {
sgerlachUnsubmitted
Done

add comment "second sample"

sgerlach: add comment "second sample"
50​
51​ QFETCH(QVector<double>, col1Data);
52​ QFETCH(QVector<double>, col2Data);
53​ QFETCH(double, tValue_expected);
54​ QFETCH(double, pValue_expected);
sgerlachUnsubmitted
Done

add space before

sgerlach: add space before
devanshuagarwalAuthorUnsubmitted
Done

Sorry, I didn't get. Add space before?

devanshuagarwal: Sorry, I didn't get. Add space before?
sgerlachUnsubmitted
Done

I meant to add an empty line to make it more readable.

sgerlach: I meant to add an empty line to make it more readable.
55​
56​ Column* col1 = new Column("col1", AbstractColumn::Numeric);
57​ Column* col2 = new Column("col2", AbstractColumn::Numeric);
58​
59​ col1->replaceValues(0, col1Data);
60​ col2->replaceValues(0, col2Data);
61​
62​ QVector<Column*> cols;
63​ cols << col1 << col2;
64​
65​ HypothesisTest tTest("Test");
66​ tTest.setColumns(cols);
67​
68​ HypothesisTest::Test test;
69​ test.type = HypothesisTest::Test::Type::TTest;
70​ test.subtype = HypothesisTest::Test::SubType::TwoSampleIndependent;
71​ test.tail = HypothesisTest::Test::Tail::Two;
72​
73​ bool categoricalVariable = true;
74​ bool equalVariance = true;
75​
76​ tTest.performTest(test, categoricalVariable, equalVariance);
77​ double tValue = tTest.statisticValue();
78​ double pValue = tTest.pValue();
79​
80​ qDebug() << "tValue is " << tValue;
81​ qDebug() << "pValue is: " << pValue;
82​ qDebug() << "tValue_expected is " << tValue_expected;
83​ qDebug() << "pValue_expected is: " << pValue_expected;
84​
85​ FuzzyCompare(tValue, tValue_expected, 0.1);
86​ FuzzyCompare(pValue, pValue_expected, 0.1);
87​}
sgerlachUnsubmitted
Done

Can you use the DEBUG() macro instead of qDebug()?
QDEBUG("tValue is " << tValue)

sgerlach: Can you use the DEBUG() macro instead of qDebug()? QDEBUG("tValue is " << tValue)
devanshuagarwalAuthorUnsubmitted
Done

QDEBUG and DEBUG macro are not working in my pc, may be due to some setting. I have to check for it. I will use the macro, once I will fix the setting.

devanshuagarwal: QDEBUG and DEBUG macro are not working in my pc, may be due to some setting. I have to check…
sgerlachUnsubmitted
Done

I thought that you already fixed it for you. If it's not working for you, please add a TODO-comment when using qDebug().

sgerlach: I thought that you already fixed it for you. If it's not working for you, please add a TODO…
88​
89​QTEST_MAIN(TTestTest)
sgerlachUnsubmitted
Done

using an absolute error of 0.01 depends heavily on tValue. Why is the error so big? Can you use relative errors instead?

sgerlach: using an absolute error of 0.01 depends heavily on tValue. Why is the error so big? Can you use…
devanshuagarwalAuthorUnsubmitted
Done

I tried using relative error. But, I am not getting what value of delta to use. If I use as small as 0.1 also then the test is not passing for values:

actual: 1.51815e-05
expected < 0.001

which is very more accurate than expected.

devanshuagarwal: I tried using relative error. But, I am not getting what value of delta to use. If I use as…
sgerlachUnsubmitted
Done

the problem here is that the expected value is not known more precise? An absolute error of 0.01 make not much sense here.
I would suggest to use a more precise expected value or test for actual < expected in this case.

sgerlach: the problem here is that the expected value is not known more precise? An absolute error of 0.
sgerlachUnsubmitted
Done

see above

sgerlach: see above
sgerlachUnsubmitted
Done

DEBUG()?

sgerlach: DEBUG()?
sgerlachUnsubmitted
Done

see above

sgerlach: see above
sgerlachUnsubmitted
Done

DEBUG()?

sgerlach: DEBUG()?
sgerlachUnsubmitted
Done

check comment

sgerlach: check comment