Tsung Fei Khang -- > > If the expectation of the empirical estimator for mean p-value is a function > of the number iterations that becomes asymptotically unbiased, then this > would explain the simulation results (attachment in the original posting), > since for small number of iterations, some bias would remain, and only > disappear (hence the "stabilization" feeling) when number of iterations is > large. > > Thanks again for all for clarifying the issue.
The issue didn't get clarified if that's the conclusion. In a permutation test each replicate is independent and equivalent to a coin toss. Thus the estimate of P is unbiased, not asymptotically unbiased. Joe ---- Joe Felsenstein [email protected] Department of Genome Sciences and Department of Biology, University of Washington, Box 355065, Seattle, WA 98195-5065 USA -- MORPHMET may be accessed via its webpage at http://www.morphometrics.org To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
