پیش بینی بازی:آیا تئوری بازی ها می تواند برنده شود؟

I present evidence to suggest that studying the use of game theory in prediction is a legitimate area of research and suggest ways in which game theory might be used to make or support predictions. Green’s study predominately assesses the accuracy of predictions by game theorists (who may have made informal use of game theory concepts) rather than predictions obtained from formal game theory models. I argue that the accuracy of predictions derived from such models is likely to be contingent on the characteristics of the conflict and provide a partial taxonomy of these characteristics, together with their hypothesised effects. I also argue that it would be worth investigating the potential use of game theory as an aid to obtaining probabilistic predictions.

The paper by Green (2002) deals with a fascinating and important area, namely the prediction of outcomes of conflicts. The paper extends earlier work, reported in Armstrong (2001), which has shown that the use of role playing to predict the outcomes of conflicts leads to forecasts that are significantly more accurate than those obtained through unaided expert judgment. In the current paper, the focus is on the relative accuracy of forecasts obtained through role-playing and game theory. This commentary addresses four questions: (i) Is it legitimate to study the use of game theory for prediction? (ii) Is game theory being assessed in the study, or game theorists? (iii) To what extent can we draw practical inferences from the results of the study. (iv) Where should future research effort be directed?

There are a number of ways in which game theory might be useful in forecasting the outcomes of conflicts. Green’s study has predominantly investigated the accuracy of the judgmental forecasts by game theory experts and found this to be inferior to the accuracy obtained through role playing, but superior to unaided forecasts produced by students. However, the use of formal game theory models to support judgment, and the use of the models in their own right, also appear to be worth investigating. In using such models there are likely to be two main sources of possible forecast error: incompleteness in the information available to the modeler and the failure of disputants to behave consistently with the assumptions of game theory. This suggests that some types of conflicts may be more amenable to predictions involving formal models than others. Further research effort could be directed into identifying these situations, while the use of game theory to produce probabilistic predictions is also worth investigating.