D21684: Hypothesis test: Dock and View for t test and z test

Devanshu Agarwal noreply at phabricator.kde.org
Sun Jun 9 08:57:35 BST 2019

devanshuagarwal created this revision.
devanshuagarwal added reviewers: sgerlach, asemke.
devanshuagarwal created this object with visibility "All Users".
devanshuagarwal added a project: LabPlot.
Herald added a project: KDE Edu.
Herald added a subscriber: kde-edu.
devanshuagarwal requested review of this revision.

  I have tried adding t-test and z test.
  For the Z Test, the backend is not done yet. In NSL library there is no defined function to get the p-value for the given z value, so I have to add my own function there.
  Tried doing model/view programming to show the summary and results via listview. This enables to give hints for every line and set edit triggers and selections to false so to get a feel of message text rather than list view.
  So available tests now, 
  Two sample independent (t and z test)
  Two sample paired (t and z test)
  one sample (t and z test)
  Sometimes, there is the case when the user wants to perform two sample tests on two columns where one column is the independent variable and the second column is having classes. So, for this, we first have to get two numeric sets of data. 
  E.g. let two columns on which user wants to perform two sample test are
  A:                      B
  class_1             value1
  class_1             value2
  class_2             value3
  class_1             value4
  class_2              value5
  Here we first have to get two sets of data like this
  class_1:             class_2:
  value1              value3
  value2              value5
  now we have to do the usual two-sample test on class_1 and class_2 columns.
  For now, only dataSourceSpreadsheet is working. After successful implementation of this we can easily shift to dataSourceDatabase

  1. Right click on spreadsheet
  2. go to "Hypothesis Test"
  3. select test from "tests" view in hypothesis dock
  4. select an alternate hypothesis radio button (two tail or one tail)
  5. change values of significance level and population mean (if you want to) else leave them to default values
  6. select columns under section independent variable on which you want to do the test. (Both columns should contain only numbers)
  7. press "Do Test" to perform the test

  R262 LabPlot



To: devanshuagarwal, sgerlach, asemke
Cc: kde-edu, #labplot, narvaez, apol
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kde-edu/attachments/20190609/87c8c919/attachment.html>

More information about the kde-edu mailing list