On Wed, Mar 2, 2016 at 11:39 PM, Alexandre Bergel <alexandre.bergel@me.com> wrote:
Hi !
- Do a public call to know what is missing for you: this is this mail :) Didier would like to work on concrete cases. I love that attitude So tell us :)
This is an excellent initiative! I personally need the following: - Shapiro-Wilk test to check for data normality - Kruskal-Wallis to analyze differences in median values between samples - Bonferroni-Dunn to support multiple comparison and controlling the group error rate - Mann-Whitney U test is a nonparametric statistical test that operates over two samples of the same size.
Great list ! At the moment, there is only Kolmogorov-Smirnov test implemented. Can you create issues in the SciSmalltalk issue tracker in order to keep track of your needs ? The issue tracker is here: https://github.com/SergeStinckwich/SciSmalltalk/issues You can also join the SciSmalltalk mailing-list: https://groups.google.com/forum/#!forum/scismalltalk Thank you Alex. -- Serge Stinckwich UCBN & UMI UMMISCO 209 (IRD/UPMC) Every DSL ends up being Smalltalk http://www.doesnotunderstand.org/