[Limdep Nlogit List] Random Parameters Frontier Model

Heath Henderson hendersonhl at gmail.com
Sat Jul 3 02:41:53 EST 2010


Hello all,

I am running a random parameters production frontier while modeling
heteroscedasticity in u.  I have come across two issues.  First, when
including Eff=u, the model spits out two different estimates of u for each
observation (i.e. two different u variables).  For any other model this is
not the case.  Second, if I try to introduce a random intercept in the
heteroscedasticity model by including One<n> in the Fcn equation I get the
following error:

Error   695: Variable in FCN=name<type> is not in RHS/RH2/HFN list.

However, if I specify One[n] the model works.  The manual says that <n>
corresponds to the heteroscedasticity model and [n] to the mean model.  This
appears to be backwards.  Any help on either of these questions would be
greatly appreciated.  My syntax is below.  Thank you!

FRONTIER ; Lhs= LN1OUTPU; Rhs=One,X; Par $
FRONTIER ; Lhs= LN1OUTPU; Rhs=One,X; Hfn=One,X2; Pds=COUNT; RPM; Fcn=One(n);
Pts=50; Halton; Par; Eff=u $

Heath

-- 
Heath Henderson
PhD Candidate, Economics
MA, International Affairs
American University
4400 Massachusetts Avenue NW
Washington, DC 20016
Tel: (612) 867-4776


More information about the Limdep mailing list