On Mon, 2007-02-19 at 09:58 -0500, Pierre Lapointe wrote:
Hello,
I have a particular situation where a single wrong observation is
impacting the results of a traditional regression to the point that
betas become unreliable. I need a way to calculate the most likely
betas. Here's an
Hi Mark,
In my example, there has been a regime change at time 25 and I'd like
to find a way to discover 1- what has changed and 2- when it did.
The problem is all that is observed are the x and y values.
unknownbetas are... unknown. If you look at x and y, you can't really
tell something has
Pierre,
Unfortunately, I don't have much in the way of hands on experience
with these, but conceptually, latent variable analysis/SEM methods seem
like they might be apropos. If so, John Fox' SEM package might be of
value here. More information here: