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]

st: Cross sectional dependence


From   Rodrigo Refoios Camejo <refoioscamejo@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Cross sectional dependence
Date   Thu, 16 Jun 2011 19:37:13 +0100

Hi all,

I have a panel with the prices of 25 drugs over 31 years (T=31 N=25).
The panel is (non-randomly) unbalanced, because it starts with very
small N (e.g. in 1980 only 4 molecules were available) and molecules
enter the market gradually and and tend to stay in the market. I am
running a model and was trying to check for cross sectional dependence
using the normal test available but none seems to work:
- xttest2 reports error: too few common observations across panel. no
observations r(2000)
- xtcsd, pesaran error: the panel is highly unbalanced. Not enough
common observations across panel to perform Pesaran's test.
insuficient observations r(2001)
- xtcd too many variables specified r(103)

I'm using xtreg a1 b1 b2 b3, fe robust

If none works, should I just assume that there is cross sectional
dependence and use the DRiscol-Kraay SE instead of the Newey-West SE?
Also, the commands available for these two assume FE pooled OLS. Is
there any way to have a RE with those corrections in place?

Many thanks in advance for the help!

RRC

*
*   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