Hedging Impermanent Loss and Optimizing Returns with ARIMA-GARCH and Options

This repository contains code and research materials for exploring a novel strategy to minimize impermanent loss and maximize returns in Uniswap V3.

• main simulation Jupyter Notebook is at src/main.ipynb

WBTC/ETH pair chosen for its high correlation and potential to reduce impermanent loss.

• Hybrid ARIMA-GARCH model forecasts pool prices and estimates volatilities.
• Walk-forward validation technique for model evaluation.

• Options-based hedging strategy to mitigate impermanent loss.
• Uses BTC-USDT and ETH-USDT options on Deribit exchange.
• Long Strangle strategy using BTC-USDT and ETH-USDT put/call options at 20/30/40% of current pool price

• "Hold On Tight" strategy with one or more boundaries, optimized using Monte Carlo simulations.
• Set up multiple trading windows within investment period based on investor preference
• Establish price boundaries based on current pool close and Futures estimates per trading window
• Implement hedging strategy and LP investment per trading window, rebalancing after every trading window

Simulation of an LP strategy with 3 windows, 1million notional and 0.95% confidence interval for price boundaries

Benchmark comparison against the HODL 50-50

• data: Contains historical Uniswap V3 pool data and options data.
• models: Contains code for ARIMA-GARCH modeling and Monte Carlo simulations.
• src: backtesting engine for simulating LP investment

1. Phase 1: Backtesting and Simulation Implementation (in progress)
2. Phase 2: Analytics Dashboard Development and Backtesting Engine Creation
3. Phase 3: Optimization

https://crypto-whales.streamlit.app

Disclaimer This project is for research and educational purposes only. It does not constitute financial advice. Investing in cryptocurrencies involves significant risks. Use any strategies at your own risk.

For more details, please refer to the full report: [link to report]