Cointegration
Two series with unit roots that share a long-run relationship, so a regression of one on the other leaves stationary errors. Only then is the regression between them meaningful.
Where you will meet it
Cointegration appears in 2 learning outcomes on the CFA syllabus. A few of them:
- Explain how to check time-series variables for cointegration and/or nonstationarity before putting them into a linear regression
“If both have unit roots, the regression is valid only if the two are cointegrated, meaning a long-term economic relationship keeps them from drifting apart and the regression errors are stationary.”
- Choose a suitable time-series model for a given investment problem and give reasons for the choice
“When two series are involved, check unit roots and cointegration.”
See also
Practise it
Practise cointegration until it is automatic: questions that explain every wrong option.
Take the free diagnostic →