Help needed! Has anyone access to StatXact? I just hacked exact two-sided p-values for rank tests (for package exactDistr, which will move to CRAN/contrib as exactRankTests soon ;-) and would like to compare the results of my implementation to that of StatXact. Could someone please calculate the exact one-sided (both greater and less) and two-sided p-values? # Data from the StatXact-4 manual, page 221 treat <- c(94, 108, 110, 90) contr <- c(80, 94, 85, 90, 90, 90, 108, 94, 78, 105, 88) My results:> wilcox.exact(x,y)Wilcoxon rank sum test data: x and y W = 52, p-value = 0.8963 alternative hypothesis: true mu is not equal to 0> wilcox.exact(x,y, alternative="l")Wilcoxon rank sum test data: x and y W = 52, p-value = 0.5656 alternative hypothesis: true mu is less than 0> wilcox.exact(x,y, alternative="g")Wilcoxon rank sum test data: x and y W = 52, p-value = 0.4636 alternative hypothesis: true mu is greater than 0 Thanks! Torsten -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Torsten Hothorn wrote:> > Help needed! > > Has anyone access to StatXact? I just hacked exact two-sided p-values for > rank tests (for package exactDistr, which will move to CRAN/contrib as > exactRankTests soon ;-) and would like to compare the results of my > implementation to that of StatXact. Could someone please calculate the > exact one-sided (both greater and less) and two-sided p-values? > > # Data from the StatXact-4 manual, page 221 > treat <- c(94, 108, 110, 90) > contr <- c(80, 94, 85, 90, 90, 90, 108, 94, 78, 105, 88) > > My results: > > > wilcox.exact(x,y) > > Wilcoxon rank sum test > > data: x and y > W = 52, p-value = 0.8963 > alternative hypothesis: true mu is not equal to 0 > > > wilcox.exact(x,y, alternative="l") > > Wilcoxon rank sum test > > data: x and y > W = 52, p-value = 0.5656 > alternative hypothesis: true mu is less than 0 > > > wilcox.exact(x,y, alternative="g") > > Wilcoxon rank sum test > > data: x and y > W = 52, p-value = 0.4636 > alternative hypothesis: true mu is greater than 0 > > Thanks! > > TorstenHi Torsten! I could do it tomorrow, when I am back in office (Stat-Xact 3 ???), if you like ... Uwe -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
> # Data from the StatXact-4 manual, page 221 > treat <- c(94, 108, 110, 90) > contr <- c(80, 94, 85, 90, 90, 90, 108, 94, 78, 105, 88) > > My results:Upps, sorry, those results where totally non-sense. I'm to tired ;-) here are the results of my current implementation for the dataset above (but data were correct!):> wilcox.exact(treat, contr)Wilcoxon rank sum test data: treat and contr W = 35, p-value = 0.0989 alternative hypothesis: true mu is not equal to 0> wilcox.exact(treat, contr, alternative="l")Wilcoxon rank sum test data: treat and contr W = 35, p-value = 0.9648 alternative hypothesis: true mu is less than 0> wilcox.exact(treat, contr, alternative="g")Wilcoxon rank sum test data: treat and contr W = 35, p-value = 0.05568 alternative hypothesis: true mu is greater than 0 Sorry again, Torsten -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Apparently Analagous Threads
- wilcox.test point estimates perverse (PR#1150)
- wilcox_test function in coin package
- Exact wilcoxon may differ in R and SPSS/StatXact (due to round off in the latter pair)
- Any interest in commercial add-on libraries based on Cyte l's StatXact/LogXact?
- Any interest in commercial add-on libraries based on Cytel's StatXact/LogXact?