> ML_ALGORITHM // AUTOREGRESSIVE-INTEGRATED-MOVING-AVERAGE-ARIMA_v1.0
ARIMA & SARIMAX (Box-Jenkins)
The definitive classical statistical framework for univariate time-series forecasting based on differencing, autoregression, and moving averages.
Classical Parametric Time Seriestime-series-forecastinghigh-intrinsicsmall (<1k)
Back to All AlgorithmsComputational Complexity
Training Complexity:O(iterations * (p + q)^2 * n) via Kalman filter likelihood
Inference Complexity:O(p + q)
Hardware Profile
CPU Friendly:Yes
Requires GPU:No
Memory Footprint:low
Interpretability & Data
Interpretability Tier:high-intrinsic
Training Data Needs:small (<1k)
Interpretability Assessment
Autoregressive (AR) and Moving Average (MA) polynomials directly quantify momentum and shock persistence.
Suitable Tasks & Supported Modalities
Suitable Tasks:
time series forecastingunivariate econometrics
Supported Modalities:
time-series
Implementing Libraries
Foundational Literature
Time Series Analysis: Forecasting and ControlGeorge E. P. Box, Gwilym M. Jenkins (1970) · Holden-Day, San Francisco
Common Pitfalls & Warnings
- Large seasonal periods (e.g., hourly data with annual seasonality s=8760) cause SARIMAX polynomial fitting to stall completely