[Limdep Nlogit List] heteroscedastic tobit model with selection

Thornburg, Steven sthornburg at wsu.edu
Tue Nov 10 03:55:04 EST 2009


Jadwiga
There is a procedure in Limdep manual V9 E27-44

namelist; x=independent variables $
namelist; z= same as independent variables unless you know what is causing heteroscedasticity $
create; y=dependent variable $
calc; kx=col(x); kz=col(z) $
tobit; lhs=y;rhs=x $	?this is the restricted homoscedastic model
calc; lr=logl; vr=s $
tobit; lhs=y; rhs=x; rh2=z; het; start=b,kz_0,vr;maxit=0 $	?This does the LM test, restricted estimates, no iterations
tobit; lhs=y; rhs=x; rh2=z; het; par ;marginal effects $	?this does the likelihood ratio test
calc; lu=logl; list;chisq=2*(lu-lr); signif=1-chi(chisq,kz) $
calc; kx1=kx+1;kxz=kx+kz$	?This does the WALD test
matrix; ghet=b(kx1:kxz) ; vghet=varb(kx1:kxz, kx1:kxz); list;waldstat=ghet'<vghet>ghet$
calc; list; signif=1-chi(waldstat,kz)$


-----Original Message-----
From: limdep-bounces at limdep.itls.usyd.edu.au [mailto:limdep-bounces at limdep.itls.usyd.edu.au] On Behalf Of dwakotki
Sent: Friday, November 06, 2009 3:25 AM
To: Limdep and Nlogit Mailing List
Subject: [Limdep Nlogit List] heteroscedastic tobit model with selection

Hello,
Can I estimate the tobit model with selection with heteroscedasticity in 
the LimDep? That is: does the following commands work:
	PROBIT; Lhs=Z; Rhs=W; Hold$
	SELECT; Lhs=Y; Rhs=X; Hfn=H; MLE; Tobit$
Is there any other way to do it easy in the LimDep (NLogit)?
Sincerely,
Jadwiga Kostrzewska

----------------------------------------------------
Oszczędzaj na kredycie.
Nie przepłacaj!
Porównaj oferty: http://klik.wp.pl/?adr=http%3A%2F%2Fcorto.www.wp.pl%2Fas%2Ffinansomat.html&sid=907


_______________________________________________
Limdep site list
Limdep at limdep.itls.usyd.edu.au
http://limdep.itls.usyd.edu.au


More information about the Limdep mailing list