Cointegration Intuition and Its Use in Pairs Trading
Summary
The document introduces stationarity as a time series property in which statistical characteristics remain stable over time, then contrasts it with nonstationary series whose means can drift. It explains that a stationary series may tend to return toward its mean after a deviation, creating a possible basis for mean-reversion strategies. The discussion is intuitive rather than a formal mathematical treatment.
Cointegration extends this idea to nonstationary series: a linear combination of them can be stationary even when each series is not. The article uses two drifting series with a stable difference to illustrate why a trader might buy the relatively cheaper asset and short the more expensive one, then reverse the positions as the spread narrows. It distinguishes cointegration from correlation and mentions unit root testing as a way to assess stationarity. The examples are conceptual; the document does not establish that relationships persist in live markets or discuss hedge estimation, costs, or practical trading risks.
Key ideas
- A stationary series has stable statistical properties, while a nonstationary series can change its long-run mean.
- Cointegration describes nonstationary series whose linear combination is stationary.
- A stationary spread between two assets can motivate a mean-reversion pairs trade.
- Cointegration is distinct from correlation, and the article mentions unit root tests for stationarity.
- The explanation is introductory and does not validate a live trading strategy.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.