Rod <[EMAIL PROTECTED]> wrote:
>I haven't looked at the data for this particular model but in general I
>would expect the errors in Y to be proportional to Y rather than
>independent.
>If this is the case linear regression is not your best estimator
>irrespective of errors in X.
Yes, in general it's worth checking into that (conditional
hetereoskedasticity). Although in this particular case there are
only 4 observations, with 2 parameters already being estimated,
so I don't think I'd try anything fancier than OLS.
Of course, given all the mistakes in the original post, like x is
not defined; probably x=t, etc., it's probably just lifted
from some problem set and we are wasting our time here... :-)
Clint
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
. http://jse.stat.ncsu.edu/ .
=================================================================
