[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Re:fitting a var...
She was correctly told that -var- does not work with panels. But VARs
are merely autoregressions of a set of y variables on a set of lags of
each of those variables. They can be efficiently estimated by OLS.
Therefore, since it is a dynamic panel data setting, she can run
xtabond2 (which I would recommend over official xtabond) for each
country/firm/etc. in the panel. xtabond2 is capable of estimating
simpler models than the Arellano-Bond, including anything one can
specify as OLS or IV.
Kit Baum, Boston College Economics
On Feb 5, 2005, at 2:33 AM, Agnieszka wrote:
I would like to fit a var to the pooled data.
I polled the data and I tried to fit the simple var command to it.
I have 3 variables and 2 lags for each.
I get the following message: repeated time values in sample.
Any idea why?
* For searches and help try: