BTC–ETH Relative-Value Trading with Cointegration and Volatility Spreads

Summary

The article proposes a relative-value approach to Bitcoin and Ethereum based on their historical relationship. Its statistical workflow measures correlation, fits a linear regression, estimates beta, tracks residuals, tests cointegration, and considers a long-short position when the residual reaches two standard deviations. It also discusses a separate options trade intended to express differences in relative volatility, alongside futures or options positions aimed at relative returns. The examples use historical market observations and specific contract illustrations from the article’s writing period.

The author argues that the pair may revert after Ethereum underperforms Bitcoin, while noting that historical correlation alone is insufficient and cointegration should be checked. The suggested setup is not guaranteed to converge: protocol-specific shocks or other extreme events could disrupt the relationship and cause substantial losses. The article stresses position sizing, but provides no out-of-sample test or evidence that the proposed trades delivered repeatable risk-adjusted returns.

Key ideas

• The proposed relative-value method models ETH against BTC and monitors regression residuals for extreme deviations.
• The article recommends checking cointegration because correlation alone does not establish mean reversion.
• A two-standard-deviation residual threshold is presented as a possible entry signal for a long-short trade.
• Options may be used to express relative volatility differences as well as relative returns.
• Asset-specific shocks can break the historical relationship, so the author highlights position sizing and tail risk.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.