A new arXiv paper analyzes over 800,000 prediction markets from Polymarket between 2023 and 2026 to predict trading volume and dispute risk at launch. The study found that market topic and contract design were strong predictors of trading volume, with interpretable features outperforming neural text models. However, accuracy significantly decreased after a platform change in listing formats, highlighting the impact of data shifts on predictive models. Dispute prediction was more stable but less precise, suitable for ranking market groups rather than forecasting individual disputes. AI
IMPACT Provides insights into how NLP models can be applied to structured financial data, though highlights limitations with data drift.
RANK_REASON Academic paper detailing a quantitative analysis of prediction market data. [lever_c_demoted from research: ic=1 ai=0.4]
- alphaXiv
- arXiv
- CatalyzeX
- Connected Papers
- DagsHub
- Gotit.pub
- Hugging Face
- Litmaps
- Polymarket
- ScienceCast
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →