Hi Jim,
The SUR method allows for contemporaneous correlation between the error terms in each equation and is more efficient than simple OLS. To consistently estimate the cross equation correlation's you need time series data for each of the dependent variables. It is true that SUR and OLS are the equivalent when each equation contains the same regressors but this is not the case when there are cross equation restrictions on the parameters. For more info see any intermediate econometrics text.
Best,
David.
--------
Sent from my HP iPAQ
-----Original Message-----
From: James Unnever <[email protected]>
Sent: Monday, June 29, 2009 2:35 PM
To: [email protected] <[email protected]>
Subject: st: seemingly unrelated regression
Need some help
