openmolecules.org

 
Home » DataWarrior » Functionality » hypothesis testing (More techniques)
hypothesis testing [message #1147] Mon, 30 November 2020 10:58 Go to next message
sansun is currently offline  sansun
Messages: 49
Registered: April 2019
Member
Is it possible to include other statistical methods to test the hypothesis?
E.g. Anova, Mann Whitney, Kruskal Wallis.
The last one is especially useful when the data is not normally distributed.

It would be nice to be able to select any of these methods to calculate p values during box plots generation.

Thanks
SS
Re: hypothesis testing [message #1149 is a reply to message #1147] Tue, 01 December 2020 17:15 Go to previous messageGo to next message
nbehrnd is currently offline  nbehrnd
Messages: 224
Registered: June 2019
Senior Member
Comment: One may interpret your suggest as a wish to extend the computation of mean
and variance (cf. http://www.openmolecules.org/help/jep.html) to offer the user a function
ready to to be called by name for descriptive statistics per column (e.g., minimal and
maximal value, lower and upper quartile; arithmetical mean, median, variance, skewness,
kurtosis); possibly followed by either a Dixon q, or a Grubbs outlier test computed
and eventually creating a visual like a q-q plot.

If so, what are other functions and statistical tests you miss in DW, e.g. in comparison
with a program like Past4 ( https://www.nhm.uio.no/english/research/infrastructure/past/)?
Re: hypothesis testing [message #1176 is a reply to message #1149] Wed, 16 December 2020 08:35 Go to previous message
sansun is currently offline  sansun
Messages: 49
Registered: April 2019
Member
Ideally it would be nice to have ability to compute all column statistics and statistical tests which can be done for e.g using Graphpad prism.

This is my wish list! I don't know how easy/difficult is to do it.
Previous Topic: Column Filtering
Next Topic: Exporting Macros
Goto Forum:
  


Current Time: Thu Dec 12 00:02:16 CET 2024

Total time taken to generate the page: 0.02379 seconds