Game theory bayesian updating
Summary This book introduces a new game theoretic equilibrium concept: Bayesian equilibrium by iterative conjectures (BEIC).
The new equilibrium concept achieves consistencies in results among different types of games that current games theory at times fails to.
• Its solution algorithm is iterative and has good computation properties.
• Can analyze more types of games than current existing games theory.
This is an "equilibrium" because the game could repeat as often as you like, perhaps even with new matchups being drawn randomly from a fixed pool for each role, and overall strategies would not change.
It's "Bayesian" because it can involve probabilistic reasoning, not only about what the other players might do, but even about what they want.
Bayesian statisticians argue that even when people have very different prior subjective probabilities, new evidence from repeated observations will tend to bring their posterior subjective probabilities closer together.
As evidence accumulates, the degree of belief in a hypothesis ought to change.A Bayesian equilibrium includes both beliefs and strategy.Everyone has some idea or ideas about what the other players may want, and they play consistently with those ideas, and the result doesn't contradict those ideas.Given its ability to typically select only a unique equilibrium in games, the BEIC approach is capable of analyzing a larger set of games than current games theory, including games with noisy inaccurate observations and games with multiple sided incomplete information games.Key Features • Provides a unified and consistent analysis of many categories of games.