A slowly decaying autocorrelation function indicates a non-stationary time series
the verdict
SUPPORTED
the evidence backs this
confidence 88/100
A slowly decaying autocorrelation function is a well-established indicator of long memory and non-stationarity in time series analysis, often modeled using fractionally integrated or integrated processes.
Evidence for · 2
ESTIMATION OF THE MEMORY PARAMETER FOR NONSTATIONARY OR NONINVERTIBLE FRACTIONALLY INTEGRATED PROCESSES
1995 · cited by 112
Paper [0] discusses fractionally integrated processes where memory parameters correspond to non-stationary behavior and long-range persistence characterized by slow decay.
See more details
More for · 1
PARTIAL AUTOCORRELATION PROPERTIES FOR NON‐STATIONARY AUTOREGRESSIVE MOVING‐AVERAGE MODELS
1992 · cited by 8
Paper [5] examines partial autocorrelation properties in non-stationary autoregressive moving-average models, noting behaviors like slow linear declines from unity.