Sorry I don't know very much about this subject, but wanted to make an issue just in case this is something that needs improving.
To be clear, currently the statictics package returns skewness=NaN when stdev=0. AFAIU there are valid arguments to return skewness=0 in this case instead.
Kurtosis is slightly more complex since there are 2 common versions Pearson/Fisher. Apparently for Pearson kurtosis=0 is a sensible default and kurtosis=-3 for Fisher.
A colleage tells me this is commonly done. I think scipy does the same: https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.skew.html
Sorry I don't know very much about this subject, but wanted to make an issue just in case this is something that needs improving.
To be clear, currently the statictics package returns skewness=NaN when stdev=0. AFAIU there are valid arguments to return skewness=0 in this case instead.
Kurtosis is slightly more complex since there are 2 common versions Pearson/Fisher. Apparently for Pearson kurtosis=0 is a sensible default and kurtosis=-3 for Fisher.