Prediction market method adapts fees to improve financial forecast consensus
Trade-Adaptive Aggregation of Probabilistic Financial Forecasts
Computational Engineering, Finance, and Science
Summary
Financial systems often predict market outcomes using information from news and reports, but combining these predictions fairly and efficiently is tricky because fees must balance rewarding useful updates without overcharging. The authors introduce a new method called SpanPM that adjusts fees based on the actual impact of each trade, improving accuracy and fairness. This method helps predictions converge faster and keeps traders’ costs lower compared to older approaches. Their experiments show SpanPM can achieve better consensus with fewer errors and lower fees in synthetic financial forecasting tests.
What this means in practice
- •For financial data teams: Implement sequential aggregation of probabilistic financial forecasts with adaptive fees to improve consensus accuracy and reduce trader costs.
- •For market design specialists: Design prediction markets with adaptive fee mechanisms that better reward informative trades without overcharging low-risk updates.
Authors
Yankai Chen, Rassul Magauin, Bowei He, Anuar Aimoldin, Sirui Song, Bin Xiao, Zangir Iklassov, Xue Liu
Abstract
Financial NLP systems produce probabilistic forecasts from news, reports, and filings. Prediction markets can aggregate these forecasts sequentially, but their fees must reward information without overcharging low-risk updates. Existing quadratic-fee mechanisms use a state-blind bound, while a local-curvature envelope remains conservative because it prices every trade at the largest permitted span. We introduce SpanPM, a prediction-market mechanism that sets the local-curvature multiplier from each trade's realized payoff spread. Its fee dominates exact Bregman exposure trade by trade, preserves no arbitrage, information incorporation, expressiveness, and bounded worst-case loss, and yields a tighter overcharge factor approaching one as trade span vanishes. Repeated global best responses converge to a common belief and become full Newton steps locally, giving quadratic rather than damped-linear convergence. We implement a deterministic bounded one-dimensional multi-basin search, audited against a dense grid. Across paired synthetic experiments, SpanPM improves 20-round consensus error by several orders of magnitude over a fixed-envelope local baseline under the same hard cap. With evolving beliefs, it preserves 96--97\% of the trader surplus achieved with exact Bregman fees while cutting excess fees by 94\% relative to the global quadratic mechanism. These results establish a trade-adaptive prediction-market mechanism for sequential aggregation of probabilistic financial forecasts.