Hi Doug
Your first model can't be right, as AM/PM is clearly a fixed effect: it has
just two levels and you are only interested in making inferences about those
two levels. So you can't have AMPM on the right of the random part of the
model. Try not to think of nesting fixed effects in random
Thanks very much Brian.
Scott
On Saturday, March 12, 2011 at 7:39 AM, Brian Inouye wrote:
In addition to Roman's suggest of the Biven's book, the first few
chapters of this other book discuss trade-offs in selecting binning
categories, and has algorithms for estimating optimal bin