Interactive Tear Sheets for Pairs Selection and Mean-Reversion Analysis
Summary
The document introduces interactive tear sheets for examining candidate trading pairs. It explains why selection requires more than a single cointegration result: Engle–Granger analysis is sensitive to which asset is treated as dependent, while Johansen analysis produces multiple vectors that merit separate review. Residual behavior, autocorrelation, and other diagnostics add further evidence for judging pair suitability.
The described module presents two analysis views: cointegration diagnostics and Ornstein–Uhlenbeck model analysis. The cointegration view gathers test results, normalized price and return plots, residual statistics, and correlation diagnostics. The OU view examines fitted mean-reverting portfolios in both asset orderings, process characteristics, simulated behavior, and adjustable inputs for exploring stopping and liquidation levels. The stated input is a dataframe of asset prices, displayed through a locally hosted interactive interface. These reports organize analysis; they do not establish that a pair will remain cointegrated, that a strategy is profitable, or that selected execution levels are suitable in live trading.
Key ideas
- Pair selection benefits from examining several tests and diagnostics together.
- Engle–Granger results can differ with the choice of dependent asset, so both orderings should be checked.
- Johansen analysis yields multiple cointegration vectors that need separate examination.
- The tear sheets combine cointegration, residual, and OU model diagnostics in interactive reports.
- The diagnostics support evaluation but do not guarantee persistent relationships or profitable trades.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.