Stationarity
Definition
In a stationary time series, certain statistical properties remain constant over time.
Stationarity in time series refers to the property that their statistical characteristics remain unchanged under time shifts; depending on the definition, different strict conditions apply.
Simply explained
Individual values may fluctuate, but their statistical pattern should not shift. Which properties must remain constant depends on the definition.
Why it matters in practice
The term helps to select, develop, evaluate AI systems in a well-founded manner and realistically assess their technical risks.
Example from business
A forecasting team checks trend and seasonal patterns before using a stationary ARMA model.
Limits and pitfalls
- An apparently stationary sample can hide structural breaks; differencing can destroy relevant long-term relationships.
Origin & history
Stationarity assumptions are fundamental in classical time series analysis and stochastic process theory.
Use in mid-sized companies
- Possible example: A forecasting team checks trend and seasonal patterns before using a stationary ARMA model.
Frequently asked questions
What does stationarity mean?
Stationarity in time series refers to the property that their statistical characteristics remain unchanged under time shifts; depending on the definition, different strict conditions apply. What should you watch out for? An apparently stationary sample can hide structural breaks; differencing can destroy relevant long-term relationships.
Sources
- Time Series · NIST SEMATECH e-Handbook, Time Series
Source checked on 10.10.2026. Editorial application examples are hypothetical; no evidence of real customer results.
Editorially reviewed on: