Learning from Inferred Foregone Payoffs

A player's knowledge of her own actions and the corresponding own payoffs may enable her to infer or form belief about what the payoffs would have been if she had played differently. In studies of low-information game settings, however, players' ex-post inferences and beliefs have been largely ignored by quantitative learning models. For games with large strategy spaces, the omission may seriously weaken the predictive power of a learning model. We propose an extended payoff assessment learning model which explicitly incorporates players' ex-post inferences and beliefs about the foregone payoffs for unplayed strategies. We use the model to explain the pricing and learning behavior observed in a Bertrand market experiment. Maximum likelihood estimation shows that the extended model organizes the data remarkably well at both aggregate level and individual level.