---
title: Seasonal ARIMA Process
url: https://www.emergentmind.com/topics/seasonal-arima-process
type: topic
---

# Seasonal ARIMA Process

A Seasonal ARIMA (SARIMA) process is a generalized class of time series models specifically designed to accommodate both nonstationary trends and periodic seasonal structures in observed data. SARIMA models leverage autoregressive (AR), moving-average (MA), and differencing operators—applied both on the nonseasonal and seasonal time scales—to transform original nonstationary series into a stationary series amenable to classical time series analysis and forecasting. Processes with fractional integration and stable innovations further extend SARIMA’s modeling flexibility, allowing for heavy tails and long-memory behavior, as encountered in finance, environmental, and engineering contexts.

## 1. Mathematical Formulation and Model Structure

The canonical SARIMA(p, d, q)×(P, D, Q)ₛ model applies both nonseasonal and seasonal differencing to a univariate time series $\{X_t\}$, with $B$ the back-shift operator ($BX_t = X_{t-1}$) and $s$ the seasonal period. The formal parametric model is

\[
\Phi(B^s)\,\phi(B)\,(1 - B)^d\,(1 - B^s)^D\,X_t
\;=\;\Theta(B^s)\,\theta(B)\,\varepsilon_t
\]

where

- $\phi(B) = 1- \phi_1 B - \dotsb - \phi_p B^p$ \hspace{1em}(nonseasonal AR)
- $\Phi(B^s) = 1 - \Phi_1 B^s - \dotsb - \Phi_P B^{Ps}$ \hspace{1em}(seasonal AR)
- $\theta(B) = 1 + \theta_1 B + \dotsb + \theta_q B^q$ \hspace{1em}(nonseasonal MA)
- $\Theta(B^s) = 1 + \Theta_1 B^s + \dotsb + \Theta_Q B^{Qs}$ \hspace{1em}(seasonal MA)
- $(1 - B)^d$, $(1 - B^s)^D$ \hspace{1em}(nonseasonal and seasonal differencing of order $d$ and $D$).

This generalizes ARIMA(p, d, q) to include multiplicative seasonal AR and MA polynomials and seasonal differencing [1211.7262][2108.05817][1212.2393][2001.08979].

## 2. Principles of Seasonal and Nonseasonal Differencing

Nonseasonal differencing of order $d$ and seasonal differencing of order $D$ are implemented via the operators $(1-B)^d$ and $(1-B^s)^D$. In spectral terms:

- $(1-B)^d$ acts as a high-pass FIR filter, amplifying mid- and high-frequency energy, with a $d$-fold zero at frequency $\omega = 0$ to remove polynomial trends.
- $(1-B^s)^D$ acts as a comb filter, introducing $D$-fold zeros at seasonal harmonics $\omega_k = 2\pi k/s$, removing periodic content of period $s$.

Impulse responses are finite:

\[
h_d[k] = (-1)^k \binom{d}{k}
\]
for $k = 0, \dotsc, d$ for ordinary differencing, and analogous expressions for seasonal differencing [1904.07632].

Standard differencing robustly removes nominal trend and seasonal periodicities, transforming the series into wide-sense stationary form, but is susceptible to over-differencing (excess attenuation and noise amplification) and frequency leakage when true spectral peaks do not align exactly with seasonal harmonics [1904.07632].

## 3. Model Identification, Estimation, and Diagnostics

Box–Jenkins methodology governs the SARIMA modeling cycle [1212.2393][2108.05817][2001.08979]:

1. **Identification**: Visual inspection and autocorrelation (ACF/PACF) analysis suggest orders $(p, d, q)$ and $(P, D, Q)$.
2. **Order Selection**: Informational criteria (AIC, BIC) and additional hypothesis testing (ADF tests for stationarity) are applied to select optimal model structure.
3. **Parameter Estimation**: Maximum likelihood methods estimate polynomials’ coefficients and innovation variance.
4. **Diagnostic Checking**: Residuals are assessed for white-noise properties and normality; Ljung–Box Q-tests and normality tests validate model adequacy.

Sparse parameterizations may be preferred—eliminating statistically non-significant terms for parsimony and stability in forecasting (e.g., SARIMA$(0,1,1)\times(4,1,0)_{12}$ for Hong Kong air traffic) [2108.05817].

## 4. Extensions: Fractional Integration and Heavy-Tailed Innovations

Fractional seasonal ARIMA (SARFIMA, ARFISMA) extend classical integer-order differencing to real-valued indices $d, D \in \mathbb{R}$, using binomial or Gegenbauer expansions to define $(1-B)^d$, $(1-B^s)^D$ [1011.5631][1211.7262]. This admits both long-memory at frequency zero and seasonal frequencies.

Innovations may follow symmetric $\alpha$-stable ($S\alpha S$) laws with characteristic function $\exp\{-|u|^\alpha\}$, $0 < \alpha \le 2$, modeling infinite-variance (“heavy-tailed”) regimes common in finance and telecommunications [1211.7262]. Stationarity and invertibility require strict constraints, e.g., $|d+D|<1/2$, $|D|<1/2$, $1<\alpha\le2$ [1011.5631][1211.7262].

Semiparametric log-periodogram approaches yield consistent and $\sqrt{m}$-normal estimators for fractional parameters in models with one or more seasonal periods [1011.5631].

## 5. Estimation Strategies: ECF, Whittle-MLE, and Simulation

Two principal estimation frameworks are dominant for seasonal fractional models:

- **Empirical Characteristic Function (ECF)**: Minimizes integrated squared deviation of the empirical joint CF over overlapping blocks to the parametric model CF. Yields simultaneous, robust, consistent, and asymptotically normal estimators for all parameters ($d, D$, AR, MA, $\alpha$) [1211.7262].
    - For ARFISMA–SαS, closed-form joint CF is available:

    \[
    c(r;\theta) = \exp\left\{ - \sum_{j=0}^\infty \bigg| \sum_{\ell=1}^{m+1} r_\ell c_{j+\ell-1} \bigg|^\alpha \right\}
    \]

- **Two-Step Method (TSM)**: First, estimates $(d, D, \phi, \theta, \Phi, \Theta)$ via MCMC Whittle likelihood in the frequency domain; then $\alpha$ via MLE on pseudo-innovations. TSM may lose efficiency when short-memory terms and stable parameters interact [1211.7262].

Monte Carlo studies demonstrate ECF’s superior finite-sample RMSE and MAE for simultaneous estimation in mixed fractional–short memory models [1211.7262].

Conditional simulation methodology enables generation of sample paths and continuation forecasts from fitted SARIMA models, as shown in R implementations; forecasts and their error structure closely agree with theoretical ARIMA predictions [1212.2393].

## 6. Computational and Methodological Advancements

Classical SARIMA is routinely implemented in statistical software via state-space Kalman filtering, conditional MLE, and recursive ARMA forecast equations [1212.2393][2001.08979]. In contrast, the univariate time-varying approach for periodic ARMA/SARIMA (Karanasos et al., 2014) eliminates recursions and high-dimensional VAR representations; analytic expressions for multi-step predictors and MSFE are constructed via continuant determinants of periodic coefficients [1403.4803]. This maintains computational tractability regardless of seasonal period.

Recent work also critiques blunt-differencing approaches in SARIMA, recommending spectral analysis and filter synthesis methods (“ARMA–SIN”) where custom-designed FIR/IIR filters precisely remove nonstationary frequencies, yielding stationary series for ARMA modeling without midband distortion [1904.07632].

## 7. Applications, Limitations, and Interpretive Dimensions

SARIMA and its fractional/stable extensions are widely deployed in economics, finance, air-traffic forecasting, environmental monitoring, and telecommunication signal analysis [2108.05817][2001.08979][1011.5631]. The ability to forecast under stable social conditions and to estimate counterfactual trajectories during crises or regime changes demonstrates SARIMA’s utility as both a predictive and impact-quantification tool [2108.05817].

Nevertheless, practical limitations, such as over-differencing, inability to completely isolate nonstationary frequency content, and sensitivity to short-memory misspecification, motivate more flexible transformation strategies and diagnostic procedures [1904.07632]. Seasonal ARIMA theory continues to evolve through advances in long-memory estimation, heavy-tailed inference, and multiseasonal generalizations.

Source: https://www.emergentmind.com/topics/seasonal-arima-process