PolymarketDataset / README.md
bbasavar's picture
Update README.md
d627935 verified
---
license: mit
task_categories:
- time-series-forecasting
language:
- en
tags:
- polymarket
- prediction-market
- finance
- trading
pretty_name: Polymarket Trading Dataset - Trades and Price Actions
size_categories:
- 10K<n<100K
---
# Polymarket Trading Dataset
## Dataset Description
This dataset contains comprehensive trading data and price action information from Polymarket, a decentralized prediction market platform. The dataset captures real-world trading activities, market dynamics, and price movements across various prediction markets.
## Dataset Contents
The dataset includes:
- **Trade Data**: Individual trades with timestamps, prices, volumes, and market identifiers
- **Price Actions**: Time-series data of price movements and market depth
- **Market Metadata**: Information about prediction markets including market questions, outcomes, and resolution criteria
- **Trading Patterns**: Historical patterns of buying and selling activities
## Use Cases
This dataset is valuable for:
- **Time-series forecasting**: Predict future price movements based on historical trading patterns
- **Market analysis**: Understand trading behavior and market dynamics in prediction markets
- **ML model training**: Train models for price prediction, anomaly detection, and market sentiment analysis
- **Financial research**: Study prediction market efficiency and trading strategies
- **Quantitative analysis**: Develop and backtest trading algorithms
## Data Format
The data is structured in a tabular format with the following key fields:
- Timestamps for all trades and price updates
- Market identifiers and outcome tokens
- Trade prices, volumes, and directions (buy/sell)
- Liquidity and market depth metrics
- User trading patterns (anonymized)
## Collection Methodology
The data was collected from the Polymarket platform using their public API, capturing real-time trading activities and price updates across multiple markets over time.
## Considerations
- The dataset reflects real-world trading on a decentralized prediction market
- Price movements are influenced by both market events and trading dynamics
- Data may contain periods of high volatility and varying liquidity conditions
- Suitable for research and educational purposes in prediction markets and trading analysis
## Citation
If you use this dataset in your research or applications, please cite:
```
@dataset{polymarket_trading_2024,
title={Polymarket Trading Dataset - Trades and Price Actions},
author={bbasavar},
year={2024},
publisher={Hugging Face}
}
```