SARIMA, which stands for Seasonal AutoRegressive Integrated Moving Average, is a time series forecasting model that extends the ARIMA model to handle seasonal patterns. The SARIMA model is particularly useful when a time series exhibits both non-seasonal and seasonal trends.
1. Seasonal Component (S): This represents the repeating pattern within each season. For example, if you have monthly data and observe a consistent pattern each year, the seasonal component captures this.
2. AutoRegressive Component (AR): Similar to ARIMA, this accounts for the relationship between a current observation and its past values. It models the linear dependence of the current value on its previous values.
3. Integrated Component (I): Denotes the number of differences needed to make the time series stationary. Stationarity is a key assumption in time series analysis.
4. Moving Average Component (MA): Like in ARIMA, this considers the relationship between the current observation and a residual term based on past observations.
The SARIMA model is denoted as SARIMA(p, d, q)(P, D, Q)s, where:
– p, d, q: Parameters for the non-seasonal component (ARIMA).
– P, D, Q: Parameters for the seasonal component.
– s: The length of the seasonal cycle (e.g., 12 for monthly data).