A Two-Dimensional Model for Game Theory Based Predictive Analytics

This article presents a two-dimensional game theory-based model for multi-actor predictive analytics. The model expands the current one-dimensional model developed by BDM to address prediction in more complex problems. So far, the single dimension models have only been able to evaluate the outcome of a problem by looking at the positions of the actors in regards to a single issue. This is while in reality, a player's strategy may depend on the dynamics of multiple issues when interacting with other players. In the first part of this work, the formulation of a two-dimensional model is presented. Then, three case studies have been included to illustrate the efficiency of the proposed tool. The presented model is explainable and can be used to predict the outcome of multilateral issues, taking into account the current dynamics of all actors with conflicting interests.