Date: Dec 15, 2012 1:48 AM
Author: Halitsky
Subject: Re plot of SEP against L

I have to do a complete re-run to get SEPs for all three of the
regressions

Ruq = c on (u,u^2)
Rub = c on (u,u*e)
Rubq = c on (e,u,u*e,u^2)

This is because I haven't yet rerun with the code I added to pick up
SEP^2 from Ivo's sigmasq, as per your post of 12/8 at 158 am ssm time.

I will do this first, so that I can present the (L,SEP) plots
alongside the n-tuples of plots which I outlined in my last post
(12/15 a 12:33 am.)