Mean reversion
The tendency of a variable to move back towards a long-run level after moving away from it. Interest rates, earnings and stationary time series show it; a random walk does not.
Where you will meet it
Mean reversion appears in 9 learning outcomes on the CFA syllabus. A few of them:
- Explain why earnings revert to the mean and how accruals change how quickly they do so
“The speed of mean reversion depends on composition: earnings with a large accrual component revert faster than earnings backed by cash, because accruals reverse and often reflect estimates.”
- Explain how a Monte Carlo simulation of forward rates works and where it is applied
“Mean reversion or bounds can keep rates in a plausible range.”
- Explain why a time series needs to be covariance stationary and what it means when a series lacks stationarity
“A stationary series has a mean-reverting level it tends back towards, while a non-stationary one does not.”
- Explain active approaches that exploit statistical arbitrage and the workings of market microstructure
“Mean reversion is the core assumption, so the key risk is a structural break that lets the spread keep widening.”
See also
Practise it
Practise mean reversion until it is automatic: questions that explain every wrong option.
Take the free diagnostic →