Non-optimal behaviour and estimation of behavioural choice models: a Monte Carlo study of risk preference estimation

摘要

Expected utility models have been extensively used in the literature to model risk behaviour and estimate producers' risk preferences where producers are assumed to maximise their expected utility in their production decisions. However, suboptimal behaviour due to budget or non-budget constraints is common, which, if not addressed, could result in biased and inconsistent estimates of parameters in behavioural choice models. A generalised model and an approximation approach are used to correct biases in the estimation of producers' risk preferences by accounting for suboptimal behaviours. We conducted a Monte Carlo simulation to evaluate the bias correction performance of the proposed model using the generalised method of moments. Our results show that the proposed model and estimation procedure produce unbiased risk preference estimates.