How can I compute the power of Dixon discordancy test in R. As there are functions for many other tests like t.test, ANOVA, proportions test etc. but not for this one. Any help will be warmly appreciated.

  • $\begingroup$ A power study using Monte Carlo simulation is described in detail at ncbi.nlm.nih.gov/pmc/articles/PMC3967640. R could reproduce this work (on a small scale, because its dixon.test calculations are slow). $\endgroup$
    – whuber
    Feb 12, 2015 at 17:03
  • $\begingroup$ Thanx @whuber for your intention. I am following the same paper which you have mentioned but there is no detail on how the power of these four test statistics are computed. Now if I programmed it on R as there is no direct function for these tests, then for power can I use the following code?? power=pnorm(TN8,lower.tail=TRUE) where TN8=max((x[n]-x[n-1])/(x[n]-x[1]),(x[2]-x[1])/(x[n]-x[1])) $\endgroup$ Feb 13, 2015 at 6:05
  • $\begingroup$ looking forward for the answer...! $\endgroup$ Feb 17, 2015 at 8:31


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.