[R] Fligner-Policello robust rank test

[Marc Schwartz (via MN)]

On Thu, 2005-09-22 at 10:44 -0400, Elisabetta Manduchi wrote:
> Can anybody tell me if there is an R implementation of the 
> Fligner-Policello robust rank test?
> Thanks,
> Elisabetta


I don't know about the Fligner-Policello test, but the Fligner-Killeen
test has been implemented in fligner.test(). From ?fligner.test:


The Fligner-Killeen (median) test has been determined in a simulation
study as one of the many tests for homogeneity of variances which is
most robust against departures from normality, see Conover, Johnson &
Johnson (1981). It is a k-sample simple linear rank which uses the ranks
of the absolute values of the centered samples and weights a(i) =
qnorm((1 + i/(n+1))/2). The version implemented here uses median
centering in each of the samples (F-K:med X^2 in the reference).


FWIW, in a Google search, I located the following SAS macro by Paul von
Hippel, which includes a comment that does not lead to optimism
regarding software implementations:

http://www.sociology.ohio-state.edu/people/ptv/macros/fligner_policello.htm

The above should not be overly difficult to convert to R, presuming you
know SAS macros...


HTH,

Marc Schwartz