Abstract
Game theory has been a useful tool to analyze the phenomena with interactive human behaviors. To use a game theory, we have to give the payoff for all players. It must be useful if we can estimate the player's payoff function from observed behavioral data, like as discrete choice model, which estimates the users' utility function from their observed behavior. This paper develops a model to identify the players' payoff function by using their observed behavioral data and the surrounding conditions which influence their choices.