Bookmark and Share

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: How to estimate SUR models with autocorrelation and heterskeasticity in panel data by Stata


From   fm gong <[email protected]>
To   [email protected]
Subject   st: How to estimate SUR models with autocorrelation and heterskeasticity in panel data by Stata
Date   Wed, 6 Jul 2011 09:56:33 -0600

Hi,

I am tring to estimate two models with correlated error terms. In
these two estimation models, each has eight independent variables,
year dummies and sector dummies. The two models have the same
dependent variables and most of the independent variables, but we do
not need to restrict the coefficients of the same independent
variables. The panel data we have are balanced. The models can be
written as follows:
y= c1+ b1 l +b2 k +b3 z +b4 s + b5 x+ b6 m +b7 z*s+ b8 m*s+ b_t year
dummies +b_j sector dummies+ error terms;
y= c2+ d1 l +d2 k +d3 z +d4 s + d5 x+ d6 n +d7 z*s + d8 n*s+ d_t year
dummies +d _j sector dummies + error terms.

I found  "sureg" in Stata 10.1 could be used to estimated the
seemingly unrelated models (SUR), but this command assumes
homoskedasticity and estimates OLS models. Our models have different
assumptions, we observed autocorrelation (AR1, PSAR1) and
heterskedasticity (HE)  in the panel data. According to some
econometrics materials, they said SUR model could accommodate the
autocorrelation and heterskedasticity, but I have not found the
command of Stata which could be use to estimate the models with AR1
and HE.  Is there "XT" serials command which could used together with
"sureg" to deal with the same situation as my models?

Could you please help me out?

Thank you so much!
Fengmei

*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index