BAYES' ESTIMATES OF THE DOUBLE HURDLE MODEL IN THE PRESENCE OF FIXED COSTS

We present a model of market adoption (participation) where the presence of non-negligible fixed costs leads to non-zero censoring of the traditional double-hurdle regression. Fixed costs arise due to household resources that must be devoted a priori to the decision to participate in the market. These costs-usually a cost of time-motivate two-step decision-making and focus attentions on the minimum-efficient scale of operations (the minimum amount of milk sales) at which market entry becomes viable. This focus, in turn, motivates a non-zero-censored Tobit regression estimated through routine application of Markov chain Monte Carlo Methods.