hypothesis testing [message #1147] |
Mon, 30 November 2020 10:58 ![Go to next message Go to next message](/forum/theme/myThemeWithHFCSS/images/down.png) |
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 message Go to previous message](/forum/theme/myThemeWithHFCSS/images/up.png) ![Go to next message Go to next message](/forum/theme/myThemeWithHFCSS/images/down.png) |
nbehrnd
Messages: 215 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 Go to previous message](/forum/theme/myThemeWithHFCSS/images/up.png) |
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.
|
|
|