Reporting Extra GG Corrected Values in rm_anova

[Esmerald1no]

I believe it would be a great addition to the package!
Generally, when reporting GG corrected p-values, the corrected F value and ddofs are also reported.
It is also a relatively easy addition.
(...)
if correction:
(...)
corr_fval = f(corr_ddof1, corr_ddof2).ppf(1-p_corr)

then during the formation of dataframe:
if not detailed:
(....)
aov["F"] = corr_fval
aov["DF"] = [corr_ddof1,corr_ddof2]
else:
(...)
aov["DF"] = [corr_ddof1,corr_ddof2]
aov["F"] = [corr_fval,np.nan]
(...)

[raphaelvallat]

Thanks @Esmerald1no for opening the issue.

I don't think we should replace the F-value and DF, because we are still reporting the uncorrected p-value. If anything, we should add the F_corr and DF_corr to the dataframe, but I am worried that it will overburden the output, which is already quite extensive.

Alternatively, we could switch to a class implementation class RepeatedMeasuresANOVA where there would be methods to report the corrected and non-corrected values. I've been wanting to rewrite some of these functions using https://github.com/matthewwardrop/formulaic as the backend, but this will take a significant amount of time which I don't have now.

Curious to hear your (and others) thoughts