[R] tests for spericity

[ripley@stats.ox.ac.uk]

I can guess you meant `sphericity', but you didn't tell us of what. Note
that the `analysis of variance' is an algorithm to partition sums of
squares, and that makes no such assumption.

I suspect that you could do what you want in lme (package nlme), if you
were able to specify it unambiguously.

On Sat, 15 Feb 2003, Corey Moffet wrote:

> Does any one know of a function/package in R that will test the assumption of
> spericity in a split-plot analysis of variance?  How are people dealing with 
> this issue in R?

-- 
Brian D. Ripley,                  ripley at stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272866 (PA)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595