Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Dynamic Panel Data
From
Clive Nicholas <[email protected]>
To
[email protected]
Subject
Re: st: Dynamic Panel Data
Date
Sun, 6 Mar 2011 22:38:40 +0000
Eric De Souza replied:
> I forgot to add the following: if the variance covariance matrix of the estimators is not correctly estimated, how can one judge the significance of the the coefficients on the LDV?
Is this another way of arguing that the VCV would be incorrectly
estimated were an LDV to be included in a FE or RE model? If so, then
the only solution is IV, as any respectable economoetrician would
argue. Frankly, sticking my head inside the oven in my kitchen and
turning the gas on would be a much more productive use of my time than
spending any more time than I have already attempting fit any more IV
models to any panel data I have collected.* In theory, they look
wonderful. In practice, they're more trouble than they're worth, even
more so if you can't find a suitable instrument which presses all the
right buttons (of which there are lots). It's little wonder that many
analysts, especially those in political science (of which I used to
include myself), choose to ignore IV and fit their panel models using
-xtpcse-.
--
Clive Nicholas
[Please DO NOT mail me personally here, but at
<[email protected]>. Please respond to contributions I make in
a list thread here. Thanks!]
*I strongly suspect that there are not one or two members of this list
who would much rather I did.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/