[Limdep Nlogit List] the best specification for the constant

[Rady Talaat]

Hello,
For the random parameters models, what is the best specification for the constant? randomising it  or fixing it? if we randomise the constant, how can we interpret the interactions between it and the experiment attributes or individual characteristics? will the model be unstable if all coefficients vary over the population?
Cheers,
Rady
_________________________________________________________________
More than messages–check out the rest of the Windows Live™.
http://www.microsoft.com/windows/windowslive/