Quantitative Finance Stack Exchange is a question and answer site for finance professionals and academics. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Assuming a directional strategy (no pairs or spread trades) is there a "standard" method for quantifying mean-reversion? Should auto-correlation, variance ratios, hurst exponent, or some other measure be preferred in all cases, or are there advantages to each given the context?

share|improve this question
up vote 7 down vote accepted

There is no standard method and many techniques can work well, including simple time series z-scoring. I'm many cases, I would recommend using the simpler approaches unless the added complexity can be justified.

However, the challenge with all techniques is the proper calibration, which is very much context sensitive. The parameter selection needs to be guided by the characteristics of the specific security and investment horizon. Furthermore, should the parameters themselves be dynamic? Nearly all processes are driven by non-stationary distributions. The amount of leverage used may also be an important consideration.

share|improve this answer

As usual, no standard method. You can maybe test for unit root over a rolling window, but as usual (again) you are going to be lagging. It will all depend on your choice of window and of the persistence of mean reversion in the market you consider.

share|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.