Stock Pairs Trading: Distance Matching and Convergence Signals
Summary
The document explains a relative value strategy that pairs stocks with similar historical price paths. It normalizes total return series, selects close matches using the sum of squared price differences, and trades the selected pairs in a later period. When the pair spread moves two standard deviations from its reference level, the strategy buys the lagging stock and shorts the leading one, then exits when prices converge. The approach assumes that temporarily separated stocks will resume moving together.
The cited research reports positive historical results, including annualized returns up to 11% in a US study and up to 15% for weekly European data. The page also discusses evidence that returns have weakened over time as pair relationships became less reliable. It notes that results can depend on market conditions, pair selection, short-selling costs, and liquidity; historical performance does not ensure future convergence.
Key ideas
- Pairs trading buys the weaker stock and shorts the stronger one when historically related prices diverge.
- The described method matches normalized stock price histories by minimizing squared differences.
- The example rule trades the closest pairs and opens positions after a two-standard-deviation divergence.
- Historical studies report positive returns, though estimates vary by market and data frequency.
- Research cited in the document finds that profitability has diminished as pair relationships became less dependable.
Tags
Cited by
- Strategies PEP/KO Heteroskedastic Filtered-Spread Re-entry
- Hypotheses PEP/KO Heteroskedastic Filtered-Spread Re-entry
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.