Model自回归分数积分移动平均 / Autoregressive Fractionally Integrated Moving Average
ARFIMA
A time-series model extending classical ARIMA by incorporating a fractional differencing operator to characterize long-memory properties, yielding autocorrelation functions that decay as a power law. It is applicable to sequences with persistent memory characteristics such as those found in finance and climatology.
Timeline (last 90 days)
Oct 8
The ARFIMA model characterizes long-memory properties through the fractional differencing operator (1-B)^d; when d>0, the autocorrelation function of the series decays as a power law k^(2d-1).
Unverified50%