This project explores the impact of transaction costs on profitability in high-frequency trading (HFT) and passive investment strategies. Using a dataset of 1-second trading data for Bitcoin (BTC), Ethereum (ETH), and Cardano (ADA) over a 2-week period, the study analyzes market microstructure, network fees, and trading strategy performance in decentralized exchanges (DEXs).
Key insights include:
- The influence of trading fees, slippage, and network costs on net profitability.
- A comparative study of active (HFT) vs passive strategies.
- The role of market microstructure dynamics and the Epps effect in trading decisions.
The study uses high-frequency cryptocurrency limit order book data sourced from Kaggle (https://www.kaggle.com/datasets/martinsn/high-frequency-crypto-limit-order-book-data) While regarding fees the dataset has been built starting from these data:
- Bitcoin (BTC): Mempool Fee Data
- Ethereum (ETH): Gas Prices
- Cardano (ADA): Explorer Fees
Key findings:
- HFT profitability is sensitive to transaction costs, favoring BTC for tighter spreads.
- Passive strategies perform better under high network congestion due to lower frequency.
- Correlation insights from the Epps effect can inform portfolio optimization.
- Clone the repository:
git clone https://github.com/andreaonorato/Transaction-Cost-Analysis-in-High-Frequency-Trading.git cd Transaction-Cost-Analysis-in-High-Frequency-Trading - Install required libraries:
pip install -r requirements.txt