[Limdep Nlogit List] Multiple Wald-tests for linear restrictions

[William Greene]

Rady. If R'VR is singular, it means that some of the restrictions are
lnearly dependent. If so, it is not possible to test all of them
simultaneously. You must remove the redundant restrictions to proceed.
/B. Greene

----- Original Message -----
From: "Rady Talaat" <rady_enviro at hotmail.com>
To: limdep at limdep.itls.usyd.edu.au
Sent: Wednesday, January 21, 2009 9:53:12 AM GMT -05:00 Columbia
Subject: [Limdep Nlogit List] Multiple Wald-tests for linear restrictions


Hello,
I am investigating the best specification for the attributes. The non-linear dummy codes were assigned to attribute by attribute but the linear effects format was retained for the remaining attributes to test whether an attribute should be specified as being either linear or non-linear. Multiple Wald-tests for linear restrictions were performed to examine the specification of the experiment attributes. However I got the following error message for some tests:
Error 310: Computing restricted least squares. R*VAR(b)*Rt singular.
What does it mean? What should I do to avoid it and perform the test?
Cheers,
Rady
 
_________________________________________________________________
Show them the way! Add maps and directions to your party invites. 
http://www.microsoft.com/windows/windowslive/events.aspx
_______________________________________________
Limdep site list
Limdep at limdep.itls.usyd.edu.au
http://limdep.itls.usyd.edu.au