Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: "Spatial" Estimated General Least Squares

From   Julia Gamas <>
Subject   st: "Spatial" Estimated General Least Squares
Date   Fri, 13 Jun 2003 02:59:01 -0700 (PDT)


I know Stata does maximum likelihood for a spatial
regression using the "spatreg" command.  The problem
is that it only does either the model with a lag OR
the model with autocorrelated errors.  However, there
is a more general model that accounts for both things:

Y = r W1 Y + XB + e
e = l W2 e + m

After testing for r and l = 0 ,I would still have to
test for whether or not there are different parameters
in different geographical areas.

Is there a command in Stata that will run this model
using Estimated General Least Squares?  So far Stata
only gives me options for time series models.  Is one
of the time series models an equivalent to my spatial
model?  Which of the time series options should I use
that would allow me to test for the different
parameters while accounting for both a lag and
correlated errors?  I am hoping I won't have to,
either enter the formulas "by hand" or switch to
another software (Matlab), since I'm pressed for time
and prefer Stata.

Any help will be greatly appreciated.


Julia A. Gamas

Do you Yahoo!?
Yahoo! Calendar - Free online calendar with sync to Outlook(TM).
*   For searches and help try:

© Copyright 1996–2019 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index