correlation-analysis

You can identify cointegrated asset pairs by scanning your asset universe for highly correlated candidates, then validating them using Engle-Granger or Johansen tests to confirm long-term statistical relationships suitable for trading.

Correlation measures short-term co-movement between assets, while cointegration identifies stable long-term relationships. This skill uses both, screening with Pearson or Spearman correlation before applying rigorous cointegration tests to validate pairs.

Dynamic hedge ratios are calculated by applying a Kalman filter to your cointegrated asset pairs. This continuously updates the ratio over time, adapting to changing market conditions better than static regression methods.

Yes, this skill performs regime-adaptive analysis to evaluate how correlations evolve across bull, bear, and high-volatility markets, ensuring your pairs trading strategy remains robust under varying market conditions.

You need pandas, numpy, scipy, and statsmodels for the statistical analysis and cointegration testing, with matplotlib and seaborn used for visualizing the resulting Z-score signals and correlation clusters.

No, high correlation alone does not guarantee a valid pairs trading opportunity. You must perform deep bivariate analysis and cointegration testing to ensure the spread is stationary before generating trading signals.