I was studying how to prepare an event related design and went through howto3.
It turns out that the design I want is very similar to that one. Therefore, it was just a matter of changing a few parameters to get what I wanted. As I looked at things more carefully it seemed to me that the whole venture is rather pointless.
Even when I changed the dt in waver to 0.2 the actual variance in norm std. dev. was relatively small. Looking at the worse one I found (10,000 iters) came back with the matrix as "VERY GOOD". It seems like it's very difficult to come up with a bad sequencing in the modelling.
However, in real life, it seems like there are "stories" about how it is very easy to come up with an identity problem in your event related design. So, is it that this modelling isn't very reflective of real life or is the fear of multicollinearity a little unwarranted as long as you randomize and jitter?