Hello,
let's say I have a linear model such as:
yt = bt*Xt + et, where et ~ iid(0,se ^2)
and I want to introduce a specification that says that the parameters are
varying over time such as:
bt = b_t-1 + ut, where ut ~ iid(0,su^2)
with su and se different.
How to estimate such a model? I am interested in the computation of b_hat.
But to be true, I am a little bit confused on how to estimate this model.
Thanks,
Nicolas
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
