Research map / Learning theory and algorithms / Prediction and decision losses
A Decomposition of Forecast Error in Prediction Markets
Miro Dudik, Sebastien Lahaie, Ryan M Rogers, Jennifer Wortman Vaughan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| decrease market liquidity b | Cost-function markets · All-security trader dynamics | convergence rate: gamma=1-Theta(b) | Hurt |
| use IND instead of LMSR at the same liquidity | Prediction markets · Market-maker equilibrium | market-maker bias: at least as large as that of LMSR, but no more than twice as large | Hurt |
| use IND instead of LMSR while matching asymptotic bias | Prediction markets · IND vs. LMSR | trades required for equal convergence error: at least half-as-many and at most twice-as-many trades as LMSR | Mixed |
| use LMSR rather than IND | Prediction markets · Markets with individually coherent trader beliefs | trades required for equal convergence error: up to a factor of two | No clear effect |
| use IND rather than LMSR at matched asymptotic bias | Prediction markets · Worst-case loss | worst-case loss: b'K log2; b log K | Hurt |
| set liquidity proportional to the total trader spending capacity | Cost-function markets · Maintaining fixed market-maker bias | market-maker bias: b proportional to N/bar a | Improved |
| use a lower liquidity setting | LMSR and IND · Combined bias and convergence error | forecast error | Mixed |
| enforce coherent prices with LMSR rather than allow independent prices with IND | Markets with individually coherent trader beliefs · Forecast accuracy | trades required for equal convergence error: up to a factor of two | No clear effect |
| increase liquidity b | LMSR · Market-maker bias | market-maker bias: bias is approximately proportional to b | Hurt |
Closest papers by meaning
- Smooth Quadratic Prediction Markets (NeurIPS 2025 · similarity 0.87)
- Adaptive Market Making with Inventory Constraints via Online Learning (AAAI 2025 · similarity 0.86)
- Hedging and Approximate Truthfulness in Traditional Forecasting Competitions (AAAI 2025 · similarity 0.85)
- A Parametric Contextual Online Learning Theory of Brokerage (ICML 2025 · similarity 0.84)
- Online Learning in Betting Markets: Profit versus Prediction (ICML 2024 · similarity 0.84)
- An Instrumental Value for Data Production and its Application to Data Pricing (ICML 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
