Measuring Bitcoin Correlation to Rank Crypto Assets and Test Relative Returns
Summary
This article studies whether crypto assets that move more closely with Bitcoin perform differently from less correlated coins. It explains Pearson correlation as a measure of linear co-movement, then describes collecting four-hour Binance futures prices for coins listed from the start of the year. The prices are normalized, and each coin's correlation with Bitcoin is estimated. The analysis forms groups of the forty most and least correlated assets and compares their average price indices as a long-short spread.
The reported result is that the higher-correlation group rose more, while shorting the lower-correlation group provided a hedge; the author says the relationship remained when correlations and returns were measured on separate periods. The article suggests that market leadership and investor attention may help explain the pattern, while low or negative correlation can reflect divergent conditions. The evidence is limited to one market and period, and the initial grouping uses future data, creating look-ahead bias. The later split reduces that issue, but the document does not detail transaction costs, survivorship effects, or risk-adjusted performance. It proposes rolling and regime-specific correlation analysis as further work.
Key ideas
- Pearson correlation measures the strength and direction of linear co-movement between asset price series.
- The study groups crypto assets by their correlation with Bitcoin and compares group price indices.
- In the reported sample, more Bitcoin-correlated assets outperformed less correlated assets, which could be used in a relative-value strategy.
- Using future observations to form groups creates look-ahead bias, though the article also reports a separate-period check.
- Rolling correlations and separate analysis of rising and falling markets are suggested extensions.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.