Distance-Based Pair Selection and Spread Reversion Trading

Summary

The distance approach selects two instruments whose historical price series moved together, then trades when their price spread exceeds a chosen threshold during a later testing period. The strategy buys the instrument with the lower price and shorts the one with the higher price, expecting their prices to converge toward previously observed levels. A common distance measure described here is the sum of squared differences between the series, associated with Gatev and coauthors’ pairs-trading work. The example figure illustrates those squared distances.

The document separates the process into pair formation, using a historical window and a distance measure, and signal generation, applying threshold rules to a testing dataset. It suggests closing when prices cross and notes extensions that trade an instrument against a weighted portfolio or compare two portfolios. This is an introductory outline rather than a full trading specification: it gives no threshold values, risk controls, transaction-cost treatment, or performance results, and convergence is an expectation rather than a guarantee.

Key ideas

• Select candidate pairs by measuring how closely their historical price series moved together.
• Use a spread threshold in a later period to trigger a long position in the lower-priced asset and a short in the higher-priced asset.
• The method expects diverged prices to converge and describes crossing prices as a possible exit condition.
• Separate historical pair selection from threshold-based signal generation on testing data.
• The framework can be extended from pairs to weighted instruments or portfolios.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.