[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Q. xtabond "goodness of fit"
Caveat 1: A week ago, I didn't know the Arellano-Bond estimator existed,
now I'm trying to use it.
Caveat 2: I'm not much of a time series guy. I haven't seen a VAR
since grad school.
Q. 1: I have two models with different specifications; they are
identical except in one I use "variable A" and in the other I use
"variable B." Both appear to perform similarly, easily passing the Wald
test. Is there a goodness of fit measure I can use to help me select
between the models?
Q. 2: The z-test on some of my estimates indicate AR(2),
which--according to the stata manuals--indicates my estimates are not
consistent. Is there a way to "quantify" the inconsistency? Better
yet, is there an accepted way to correct it?
Thanks to all in advance!
Noel D. Campbell
Assoc. Prof. Econ.
U. Central Arkansas
* For searches and help try: