efficiency (partial) balance

[Peter Cameron]

On Tue, Nov 04, 2003 at 11:44:33AM +0000, R A Bailey wrote:
> This message concerns efficiency factors, but an immediate conclusion
> is that we should say nothing about statistical properties of
> unequally replicated designs in the forthcoming release.
> 
> We can get away with being sloppy for efficiency balance.
> But if we want to extend it to some idea of
> efficiency partial balance then I think we need to start with the
> matrix N^{-}LN^{-}.
> 
In fact we don't have an ything about partial balance in the current
release - this is one of the jobs needed for the next release. I think
your comments are more relevant for the Tehran paper, where the deadline
is not quite so imminent, so I'd like to see this discussed at leisure.

Peter.