Random Choice and Learning

Context-dependent individual choice challenges the principle of utility maximization. I explain context dependence as the optimal response of an imperfectly informed agent to the ease of comparison of the options. I introduce a discrete choice model, the Bayesian probit, which allows the analyst to identify stable preferences from context-dependent choice data. My model accommodates observed behavioral phenomena—including the attraction and compromise effects—that lie beyond the scope of any random utility model. I use data from frog mating choices to illustrate how the model can outperform the random utility framework in goodness of fit and out-of-sample prediction.