Suboptimality of Probability Matching − a Formal Proof, a Graphical Analysis and an Impulse Balance Interpretation

The objective of the paper is to study how the tax burden arising from an exogenWe prove suboptimality of probability matching in prediction tasks with an arbitrary (finite) number of outcomes and repetitions. For the popular case of binary prediction tasks, we also provide a graphical representation of the result. Finally, we relate probability matching to impulse balance equilibrium theory and show when probability matching is consistent with its predictions.