Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: RE: constant variable as IV in panel?


From   Matthias Opfinger <opfinger@vwl.uni-hannover.de>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: RE: constant variable as IV in panel?
Date   Thu, 23 Jun 2011 14:47:11 +0200

Of course, you're right. I confused some things here. The climatic variable was a useful instrument for some other project I was working on where ethnic polarization had to be instrumented. I'm sorry for that confusion. But still my question remains. What do I do if I have a panel but only one observation for the possible instrument, e.g. from the past and it doesn't up to today. Is it somehow still possible to use panel iv methods?

Matthias

Am 23.06.2011 14:37, schrieb Austin Nichols:
Matthias Opfinger<opfinger@vwl.uni-hannover.de>:
What does "national identity" mean here?  What is the outcome variable?
I find it hard to believe the exclusion restriction for any
climate-based measure, in most cases
[but see e.g. http://www.nber.org/papers/w14031 for a good example
from economics].
It is in the nature of climate to affect everything directly,
including the outcome, even if only in a small way.

On Thu, Jun 23, 2011 at 5:52 AM, Matthias Opfinger
<opfinger@vwl.uni-hannover.de>  wrote:
Thanks for the remark. Actually I try to instrument national identity. The
instrument I want to use is the percentage of the land mass in temperate
climatic zones. In a cross section that worked quite well, but I'm not sure
how to use this in a panel setting.

Am 23.06.2011 11:36, schrieb Nick Cox:
There are many instrumental variable experts on the list who are better
placed than I to comment on that side.

I just want to point out that the implicit climatology here is fallacious,
not that it needs a geographer who has occasionally published on climate to
point it out. Whatever scheme of climatic zones is being considered here,
and there are many standards to choose from,

1. it won't be true for most if not all large countries that they fall in
just one climatic zone (think Russia, China, Brasil, USA, etc.)

2. it won't even be true for many smaller countries

3. the assumption that the climate is unchanging is also incorrect,
although probably less important than 1 or 2 for this study.

More positively, climatic zonation throws away information that is encoded
in variables like precipitation, temperature, etc. but using the latter will
underline the heterogeneity problem identified in 1 and 2.

Nick
n.j.cox@durham.ac.uk

Matthias Opfinger

I want to estimate a panel model with observations on approx. 90
countries for five points in time between 1980 and 2005. However, one of
my explanatory variables is endogeneous. The only possible instrument is
a variable on climatic zones. So it would be the same for each country
over all points in time. How can I put this into my model? Do I just put
the same value for each country over all the points in time and then use
xtivreg?
*
*   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/


--
Dipl.-Ök. Matthias Opfinger	

Leibniz Universität Hannover	
Institut für Makroökonomik
Department of Economics

Koenigsworther Platz 1
30167 Hannover
Germany

Phone: +49-(0)511-762-5655
http://kaldor.vwl.uni-hannover.de/

*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index