Scoring Rules for Performative Binary Prediction
arXiv:2207.02847v13 citationsh-index: 4
Originality Highly original
AI Analysis
This addresses a fundamental issue in prediction markets and expert systems where incentives may lead to undesirable world manipulation, offering a novel solution.
The paper tackles the problem of performative prediction where expert forecasts can influence outcomes, showing that standard proper scoring rules can incentivize manipulation, and proposes a new class of scoring rules to prevent this.
We construct a model of expert prediction where predictions can influence the state of the world. Under this model, we show through theoretical and numerical results that proper scoring rules can incentivize experts to manipulate the world with their predictions. We also construct a simple class of scoring rules that avoids this problem.