FTB-C000315 / Statistical method

Recursive Exponential Weighting

Recursive Exponential Weighting updates an exponential smoother from the current observation and the previous smoothed state.

Also known asrecursive EWM

Definitions

In plain terms

Its standard recurrence is alpha times current input plus one minus alpha times the prior state.

Technical

The contract must define alpha, seed, first-ready output, missing observations, resets, order, and numerical precision.

Scope

It can differ from normalized adjusted weighting in early observations.

Formula

E_t = alpha*x_t + (1-alpha)*E_(t-1)
LaTeX: E_t=\alpha x_t+(1-\alpha)E_{t-1}
SymbolMeaningUnit
alphasmoothing constantratio
x_tcurrent observationinput unit
E_(t-1)prior smoothed stateinput unit

Output unit: input unit

Examples

  • A governed lesson uses Recursive Exponential Weighting only with declared inputs, timing, parameters, and edge-case behavior.

Common misconceptions

  • It can differ from normalized adjusted weighting in early observations.

Concept relationships

Where this concept is used

Evidence and governance

  1. Moving Average and Smoothing Methods NIST/SEMATECH · official standard

    Supports: preferred label, short definition, technical definition, formula

    Limits: Forecasting notation and initialization can differ from technical-indicator conventions and must be translated explicitly.

Reviewed by
fintech-builder-batch-006
Last reviewed
2026-07-27
Next review
2027-07-27
Record status
evidence reviewed

This record is evidence-reviewed and readable, but not yet promoted to published — it is served noindex,follow and excluded from the sitemap.