Testing and Backtesting Mean-Reverting Crypto Perpetual Pairs
Summary
This project outlines a statistical arbitrage approach to trading cryptocurrency perpetual contracts on Binance. It screens contract price series for stationarity and cointegration, then forms a spread between a selected pair and uses deviations from the spread’s mean as entry and exit signals. Perpetual funding payments are described as a mechanism that helps keep contract prices anchored to spot. The project also discusses selecting parameters through historical backtesting.
The reported screening found a small subset of stationary series among the contracts examined, and ETC-USDT and RLC-USDT were chosen as an example pair. The backtest reports a Sharpe ratio of 1.14 and a final portfolio value of $90,717.54 from a $10,000 starting value. These are historical results from the project, not evidence of live performance. The text cautions that pair trades can lose money, contracts can be delisted, spread coefficients matter, and market regimes change, so parameters and recent data require review. Some methodology and results are omitted in the supplied text, limiting independent assessment of the tests and execution assumptions.
Key ideas
- A price spread between related contracts may be more mean reverting than either contract price on its own.
- Stationarity and cointegration tests are used to screen candidate pairs before constructing a spread.
- The strategy opens positions at extreme spread levels and closes them nearer the estimated mean.
- Perpetual funding transfers between longs and shorts and is intended to help align contract and spot prices.
- Backtest results are historical and depend on parameter choices, market conditions, and execution assumptions.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.