the definition for d is

d, the magnitude of the investigated effect relative to σ, the standard deviation of the sample

which sounds like effect size, for example mean difference divided by sigma

However, the examples below use d for the original magnitude (mean difference) and explicitly divide by sigma, AFAICS

maybe introducing another variable for the original magnitude (difference of mean from null value)

R has both kinds of functions (those that use original magnitude and sigma, power.t.test in package stats, and those that use effect size d, package pwr)

Incorrect file path to data files specified in exercises

• swim100.csv (statsPython.rst)
• Table 2.8 Waist loss.xls (statsPython.rst)
• babyboom.dat.txt (statsPython.rst))

First of all, this is a phenomenal resource - thank you so much for putting this together. Regarding the license, does this mean that this repo is no longer open to contributions? Also, what license have you switched to? Thanks.

Please add the Kolmogorov-Smirnof Test.

I would love to see a section on parameters that describe a distribution such as:

• Shape: http://en.wikipedia.org/wiki/Shape_parameter
• Scale: http://en.wikipedia.org/wiki/Scale_parameter

So the research question is:
I have some data --> how do I know the right distribution model which best describes my data?

http://work.thaslwanter.at/Stats/html/statsDistributions.html#probability-and-samples

says:

Survival function (SF): 1-CDF

Please add a explanation.
Especially why it is called survival and what it tells on the data.

Hi,

Could you add in the readme file, the links to the compiled version (http://work.thaslwanter.at/Stats/html/) and the pdf?

Could you also add a link in the document (near the license for example) to this repository?

That would be really useful for readers, contributors etc.

Thanks!