[Limdep Nlogit List] Problems implementing Herriges and Shogren anchoring approach

[Oliver Frör]

Hello,
 
in the context of estimating willingness-to-pay using double-bounded
dichotomous choice data I am trying to implement the anchoring model by
Herriges and Shogren (1996) where the response to the follow-up bid is based
on an anchoring-adjusted WTP function WTP2= (1-gamma)*WTP1 + gamma*bid1. For
the estimation in LIMDEP I am using the MAXIMIZE command where three
different probability functions for the response to the first bid, the high
and the low follow-up bid are specified for the Log-likelihood function,
either in a probit or logit form.
I had trouble specifying the probabilities for the follow-up questions
correctly since "gamma" enters the function twice. For the estimation I
tried to use two different labels f1 (as a parameter for WTP1) and f2
(parameter for bid1) and restrict f2 to equal (1-f1). However, the
respective restriction option "; Rst" did not seem to work and on top I
receive an error message (Logminus). Is this the correct approach? Do you
have any alternative approaches I could try? I would be grateful for any
help in this case.
 
The full citation of the approach is: Herriges, J.A., Shogren, J.F. (1996):
Starting Point Bias in Dichotomous Choice Valuation wit Follow-Up
Questioning. Journal of Environmental Economics and Management 30, 112-131.
 
Thanks and regards,
 
Oliver Froer
 
University of Hohenheim, Germany
froer at uni-hohenheim.de