correlation-analysis

Crypto correlation analysis computes cross-asset correlation matrices from return series using Pearson, Spearman, and Kendall methods, quantifying diversification quality and market-factor dominance to support portfolio risk assessment.

Calculate rolling correlation using z-score logic on two-asset return series to detect regime shifts and correlation breakdowns. This identifies structural changes in co-movement critical for reliable pairs trading decisions.

Tail dependence estimates how crypto assets move together during extreme market downturns. By analyzing crash co-movement from return series, it reveals stress-aware correlation structure that standard Pearson correlation often misses.

Yes, hierarchical clustering groups crypto assets by deriving clusters from correlation matrices. This structure discovery process identifies effective diversification levels and reveals market-factor dominance across your portfolio.

Yes, running cross-asset correlation analysis requires Python execution with pandas, numpy, and scipy for matrix computations and clustering. It optionally uses httpx to fetch real crypto price data via CoinGecko.

Using return series instead of price levels prevents spurious correlation from non-stationary price trends. Computing Pearson, Spearman, and Kendall correlations on returns accurately characterizes true co-movement for diversification metrics.