These tests are in any case not always very
informative and may constitute at best an
ornamental oddity in a cul de sac, or something
required ritually by advisors or reviewers.
There are at least two alternatives:
1. Pretend that your data are survival times
and fire up -streg- to fit a three-parameter lognormal.
2. Take logs and fire up -qnorm-. This is not
quite the same but it is often helpful.
Nick
[email protected]
Salas, Jesus M.
> I believe it is fail to reject because the z statistic is not
> significant at any level...
Jiehua CHEN
> Does anyone know how to interpret the result of swilk
> function. Like my
> result is
> swilk ag, lnnormal
>
> Shapiro-Wilk W test for 3-parameter lognormal data
> Variable | Obs W V z Prob>z
> -------------+-------------------------------------------------
> ag | 1199 0.79999 148.830 -1.088 0.86173
>
>
> Shall I reject or accept the null hypothesis?
*
* 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/
