There is a good reason for this. The
P-value is calculated internally
as 2 * normprob(-abs(`z')) where `z'
is the statistic that is reported as r(z).
That is, the mapping is one to one, and
you lose no information from not having
the P-value returned as well.
So after -ranksum- you just need to calculate
2 * normprob(-abs(r(z))).
Nick
[email protected]
Herve STOLOWY
> I perform a ranksum test (i.e. Wilcoxon - Mann-Whitney U test). With
> -return list-, I get the saved results. Unfortunately, I see
> the (z) but
> not the Prob(z) in this list.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
