Correlation, Cointegration, and the Risks of Pair Trading

Summary

The document raises a conceptual question about pair trading. The strategy described is to short the relatively expensive asset and buy the relatively cheap one when the spread between them is expected to revert toward its mean. The author asks whether strong positive return correlation and cointegration undermine that trade, since both asset prices might rise while one rises faster, potentially causing losses on the paired positions even as the spread moves toward its expected level.

This is useful as a statement of the distinction a pairs trader must examine: co-movement in returns, a long-run relationship between price series, and the behavior of the traded spread are related but not interchangeable. However, the document contains no answer, empirical evidence, entry or exit rules, or discussion of hedge ratios and position sizing. It therefore frames a strategy-design concern without establishing whether the proposed scenario invalidates a pair trade. Further analysis would need to specify how the spread is constructed and what return or risk objective defines success.

Key ideas

• Pair trading seeks to profit from relative mispricing as a spread returns toward its mean.
• Strong return correlation and cointegration do not by themselves specify how a spread trade will perform.
• Both assets can rise while their relative prices move in a way that affects the paired position.
• A useful analysis must define the spread, hedge ratio, and position sizing.
• The document poses the issue but supplies no answer or empirical evidence.

Tags

# Pair trading with strong positive correlation in asset return and cointegration


# Pair trading with strong positive correlation in asset return and cointegration












I am a newbie exploring pairs trading. The main idea is to short-sell the relatively overvalued stocks and buy the relatively undervalued stocks based on the mean-reverting properties of the spread between the asset pair.

However, if the asset pair has strong positive correlation in returns and are also cointegrated, wouldn't pair trading be a bad idea in this case? For instance, when the spread is way below the mean and it is expected to increase back to its mean, it could be that the price of asset A is increasing much faster than the price of asset B, while both asset prices are increasing. In that case, even though the spread between the 2 asset will widen, following the pair trading strategy would cannibalise the profit.

Am i mis-understanding/missing certain criteria about pair trading?

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

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