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]
st: xtsur-Using SUR in an unblanaced panel data, cross equation restrictions, clustering and heteroscedasticity
From
"Abdalla, Ahmed" <[email protected]>
To
"[email protected]" <[email protected]>
Subject
st: xtsur-Using SUR in an unblanaced panel data, cross equation restrictions, clustering and heteroscedasticity
Date
Mon, 25 Nov 2013 09:48:28 +0000
Dear Statalist
I have a system of 5 equations and I want to test cross equations restrictions. As far as I understand, I have to use SUR if I seek efficiency and to be able to test the restrictions that involve more than one equation. Stata has the code sureg that tests for SUR in a cross section. In order to test for cross equation restrictions using a panel data set, stata has the code xtsur that is based on an approach originally developed by Brion (2004) for one way random effect estimation of SUR in unbalanced panel set.
My concerns are:
1- If I doubt heteroscedasticity which is always the case using financial data, can I use any heteroscedasticity-robust standard errors ?
2- Can I cluster standard error by firm or time ?
3- xtsur deals with random effect estimation of SUR. What if I test the models using Hausman test and find that the fixed effect model is more appropriate, can I still using SUR in this setting in panel data? code?
4- Is there any alternative other than SUR here in STATA in order to test cross equation restrictions in a panel dataset ?
Thanks
Ahmed
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/