Recreation demand models are typically plagued by limited information on site attributes.
If these unobserved site attributes are correlated with the observed characteristics and/or the
travel cost variable, the resulting parameter estimates are likely to be biased. We develop
a Bayesian approach to estimating a model that incorporates a full set of alternative-speci c
constants, insulating the key travel cost parameter from the in
uence of unobservables. The
proposed posterior simulator can be used in the mixed logit framework in which some parameters
of the conditional utility function are random. We apply the estimation procedures to data from
the Iowa Lakes Project.