Data & Analytics
(Exponentielle Glättung)Exponential Smoothing
Also known as:
ETS
Holt-Winters
Exponential Smoothing State Space
Updated: 2/11/2026
A family of statistical time series methods that exponentially weights current observations more heavily than past ones.
Quick Summary
ETS/Holt-Winters weights recent data more heavily – simple, fast, and surprisingly accurate as a forecasting baseline.
Explanation
Simple (level), Double/Holt (level + trend), Triple/Holt-Winters (level + trend + seasonality).
Marketing Relevance
Often surprisingly accurate and faster than complex ML models. Standard baseline.
Common Pitfalls
Only univariate. Cannot natively model multiple seasonalities.
Origin & History
Brown (1956), Holt (1957), Winters (1960). ETS framework by Hyndman et al. (2002) unified all variants.
Comparisons & Differences
Exponential Smoothing vs. ARIMA
ETS decomposes into level/trend/season; ARIMA models autocorrelations.