Predicting strategic medical choices: An application of a quantal response equilibrium choice model

Quantal response equilibrium choice (QREC) models are structural behavioral models that account for bounded rationality and strategic interactions in analyses of games where each player’s payoff is a vector. We revisit the question of how market competition affects pro-social behavior and fit a QREC model to data from an incentivized laboratory experiment, where the participants make decisions on medical treatment for abstract patients in monopoly, duopoly and quadropoly games. Our results demonstrate that competition can cause substantial behavioral responses without any changes in pro-social preferences if one allows for the possibility that competition influences the degree of randomness in decision-making. We find that a QREC model with fixed preference parameters provides precise out-of-sample predictions of behavior in games with vector payoffs. A Monte Carlo study is performed to show that the two-step estimator is accurate.