KomFi. Academy

Exponentially weighted moving average (EWMA)

FRM Part 1 glossary · Quantitative Analysis

Definition

A volatility-estimation model that updates variance as σ^2ₜ = λσ^2ₜ₋_1 + (1-λ)r^2ₜ₋_1, weighting recent observations more heavily via the decay factor λ. It is the GARCH(1,1) special case with ω = 0 and α + β = 1, so it does not mean-revert.

Exponentially weighted moving average (EWMA) in plain English

EWMA is a way to estimate how jumpy (volatile) a price is right now, giving recent days more say than old ones. Each day you update your estimate by blending yesterday's estimate with yesterday's actual move, controlled by a "decay factor" λ between 0 and 1: σ^2_t = λ,σ^2_t-1 + (1-λ),r^2_t-1. A high λ (like 0.94) means memory fades slowly, so old days still count a bit, while the newest day always gets the freshest weight — like a running average that quietly forgets the distant past. One quirk: unlike some fancier models, EWMA has no built-in "normal" level to drift back toward, so it just follows the recent action without assuming calm will automatically return.

Also called: EWMA

Learn these first

Practice Exponentially weighted moving average (EWMA)

Within Quant — Time Series & Volatility Modeling, which term matches this definition: “volatility estimate that applies…

Sign up free: all 170 FRM Part 1 terms as flashcards, plus practice questions with full explanations →

Related FRM Part 1 terms

FRM Part 1 practice questions · FRM Part 1 flashcards · FRM Part 1 study guide