s-news
[Top] [All Lists]

Bug in qwilcox() ?

To: s-news-list <s-news@lists.biostat.wustl.edu>
Subject: Bug in qwilcox() ?
From: Gerrit Eichner <Gerrit.Eichner@math.uni-giessen.de>
Date: Tue, 24 Jun 2003 12:34:49 +0200 (CET)
Dear list,

I've encountered the following strange behaviour when computing a quantile
of Wilcoxons rank sum distribution using qwilcox():

> qwilcox( 0.95, 36, 36)
Problem in cut.default(x, ..1): breaks must be given in ascending order
and contain no NA's
Use traceback() to see the call stack


Similar sample sizes (even or odd, and at least the ones, I've tried)
don't seem to present any problem, e. g.:

> qwilcox( 0.95, 34, 34)
[1] 1307
> qwilcox( 0.95, 35, 35)
[1] 1383
> qwilcox( 0.95, 37, 37)
[1] 1540
> qwilcox( 0.95, 38, 38)
[1] 1621


I've checked the archive, but didn't find anything. Does anybody have an
idea about what is happening above, and even better: how to work around
this problem (bug?) ? Thank you.

 Best regards  --  Gerrit

PS: S-PLUS Version 6.0 Release 1 for Sun SPARC, SunOS 5.6 :  2000.

 -------------------------------------------------------------------------
  Dr. Gerrit Eichner                        Mathematical Institute of the
  gerrit.eichner@math.uni-giessen.de        Justus-Liebig-Univ. Giessen
  Tel: +49-(0)641-99-32104            Arndtstr. 2, 35392 Giessen, Germany
  Fax: +49-(0)641-99-32029      http://www.math.uni-giessen.de/Stochastik
 -------------------------------------------------------------------------


<Prev in Thread] Current Thread [Next in Thread>