Risk identification in traffic is mandatory to ensure safe driving for automated vehicles. Dangerous scenarios could result from an inaccurate estimation of the future trajectories of the other traffic participants, under the uncertainty of the human behavior. Prediction is therefore necessary in order to guarantee a safe decision making. Nevertheless a reliable estimate can not be limited to a short term prediction based on dynamic or kinematic models, but it has to take into account long-term interactions and influence between the traffic participants in the scenario. This article presents a methodology to predict the future trajectories of vehicles in a traffic environment, considering the information coming from sensors in terms of actual state of the vehicles, but also interactions and mutual influence between them, that can come from possible future traffic outcomes. The methodology used is the Bayes´ Theorem, it proposes a game-theoretic framework in the form of a Mixed Strategy Nash Equilibrium (MSNE) as prior knowledge to model the reciprocal influence between traffic participants, while data coming from sensors correct the prior estimate.