Skip to content
All library documents

Inferring Mean-Reversion Speed from a Deterministic Terminal Value

Article Quant Q&A · Author: Lisa Ann

Summary

The document asks how to infer the mean-reversion rate in a deterministic process that moves toward a known long-run level. Its stated dynamics contain only a drift term proportional to the distance from that level, with no random component. The terminal observation is described as being approximately equal to the long-run value at a specified finite time.

The author asks whether a closed-form estimate or proxy for the speed parameter can be obtained from that condition. No derivation, estimator, numerical example, or evidence is provided, so the document frames an identification problem rather than resolving it. A practical estimate would also require a precise tolerance for “approximately,” along with the starting value and observation horizon; convergence toward the target alone does not specify a unique finite-time speed.

Key ideas

  • The process approaches a known long-run level at a rate controlled by a mean-reversion parameter.
  • The setup is deterministic because it has no diffusion term.
  • The terminal value is only approximately equal to the long-run level at a finite horizon.
  • The document asks for a closed-form estimate or proxy but provides no solution.

Tags

Full text
# Exact value of mean reversion rate knowing terminal value of the process


# Exact value of mean reversion rate knowing terminal value of the process












Let you have the following mean reverting process:

$\text{d}x_{t}=a(\theta-x_{t})\text{d}t$,

where the diffusion term is absent, that is this process is not stochastic.

Let you know the value of $\theta$.

You also know that at time $t=T$ it must be $x_{T}\simeq\theta$.

(That is when $|x_{T}-\theta|$ is so small to be negligible because $x_{t}=\theta$ when $t\rightarrow\infty$).

Does any closed form and/or a proxy of $a(\theta,T)$ exist?

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.