Using Cointegration to Build Mean-Reverting Trading Portfolios
Summary
This introduction explains how cointegration can help create a mean-reverting portfolio from price series that are not themselves mean-reverting. By combining multiple assets with suitable weights, a trader may construct a spread or portfolio whose value tends to return toward an average. The module described includes tools for testing whether a given set of series can form such a portfolio, estimating the combination, and generating trading signals from the resulting portfolio-price series.
The discussion motivates the approach by contrasting financial prices with naturally mean-reverting processes, illustrated by historical Nile River levels. It gives no empirical trading results or detailed estimation procedure in this introduction, and it does not establish that a candidate portfolio will remain stable or profitable. The scope is time-series mean reversion through cointegrating combinations; cross-sectional mean reversion based on relative cumulative returns within a basket is explicitly outside the module’s coverage.
Key ideas
- Most individual financial price series are not assumed to be mean-reverting.
- Cointegrating combinations of multiple assets can produce a portfolio series with mean-reverting behavior.
- The described tools test candidate asset sets and identify portfolio combinations.
- Trading signals can be generated from the constructed portfolio-price series.
- Cross-sectional mean reversion is outside the scope described here.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.