What problem does it solve?
This Skill helps you discover and validate assets with stable co-movement relationships, then turn that evidence into actionable pair-trading signals through correlation, cointegration, and spread mean-reversion analysis.
Core Features & Use Cases
- Co-movement Discovery: Scan a universe to rank candidates by Pearson/Spearman correlation and filter to a Top-K pool for further testing.
- Deep Return-Correlation Analysis: Quantify bivariate relationships with multiple correlation measures, OLS beta/alpha/R², rolling correlation, and spread Z-score features.
- Cointegration & Spread Diagnostics: Run Engle-Granger and Johansen tests, compute spread half-life, and estimate both static and dynamic (Kalman) hedge ratios.
- Realized Correlation & Regime Insight: Measure how correlations change across bull/bear/high-volatility regimes using rolling windows and conditional summaries.
- Pair-Trading Signal Generation: Convert cointegration-validated spreads into entry/exit/stop Z-score state-machine signals for long/short spread positioning.
Quick Start
Run correlation and cointegration screening for your target asset, then generate pair-trading long/short signals using Z-score thresholds, half-life-informed lookbacks, and (optionally) a Kalman dynamic hedge ratio.
