Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Econometrics›Robust SARIMA Model
Regression modelEconometrics / time series

Robust SARIMA Model

Robust Seasonal Autoregressive Integrated Moving Average Model · Also known as: robust SARIMA, outlier-resistant SARIMA, robust seasonal ARIMA, M-estimator SARIMA

Robust SARIMA extends the classical Seasonal ARIMA framework by replacing the standard least-squares criterion with a robust loss function — such as an M-estimator — so that outliers and heavy-tailed innovations in seasonal time series cannot distort parameter estimates or invalidate forecasts.

ScholarGate
  1. Regression model
  2. v1
  3. 2 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Robust SARIMA model
ARIMA modelRobust RegressionSARIMA modelX-13ARIMA-SEATS

When to use it

Use Robust SARIMA when you have a seasonal time series that likely contains outliers, measurement errors, or occasional structural shocks — common in macroeconomic indicators, energy demand, and retail sales data. It is especially valuable when forecasting accuracy matters more than a single anomaly, or when standard SARIMA diagnostics reveal non-normal, heavy-tailed residuals. Do not use it when the series is clean and Gaussian: the robustness comes at a modest efficiency cost, and classical SARIMA is then slightly more efficient. Also avoid it when you need a quick, interpretable baseline; the added complexity of the robust iteration requires more careful software handling.

Strengths & limitations

Strengths
  • Resists distortion from additive outliers, innovative outliers, and level shifts without requiring manual data cleaning.
  • Delivers reliable seasonal and trend decomposition even in contaminated series.
  • Parameter estimates and forecasts remain close to those you would obtain from the clean data.
  • Applicable to the full range of SARIMA orders, making it a drop-in robustified alternative to the classical model.
  • Formal M-estimation foundation provides asymptotic theory and valid inference under mild contamination.
Limitations
  • Estimation is iterative and computationally heavier than classical SARIMA, which can matter for high-frequency long series.
  • Implementation requires specialised software (e.g., R packages robets, robustarima, or custom M-estimator code); not available in standard econometric suites out of the box.
  • Modest efficiency loss relative to classical SARIMA when the data are actually Gaussian and outlier-free.
  • Model selection (order identification via ACF/PACF) is still based on the structure of the data and can be complicated by contamination before robust estimation is applied.
  • Distinguishing additive from innovational outliers in practice requires diagnostic care and can affect the cleaning strategy.

Frequently asked

How does Robust SARIMA differ from simply removing outliers before fitting standard SARIMA?

Manual removal requires knowing which observations are outliers in advance and risks deleting real signal. Robust SARIMA down-weights potential outliers automatically during estimation, using formal influence functions, so the threshold is data-driven and the procedure is part of the model fit rather than a pre-processing judgement call.

Which ψ function should I choose — Huber or Tukey bisquare?

Huber's function provides a good balance: it is quadratic for small residuals (like OLS) and linear for large ones, ensuring continuity. Tukey's bisquare redescends to zero for very large residuals, offering stronger resistance to extreme outliers but at the cost of a non-convex objective. For most economic series, Huber is a safe default; use bisquare when outliers are severe.

Can I use standard AIC/BIC for order selection in Robust SARIMA?

Standard information criteria based on Gaussian likelihood are distorted by outliers. Prefer robust model-selection criteria (e.g., robust AIC based on M-estimator log-likelihood) or use the ACF/PACF of robust residuals after a preliminary contamination-resistant fit.

Does Robust SARIMA handle all four types of outliers?

Most implementations directly address additive outliers (AO) and innovational outliers (IO). Level shifts (LS) and temporary changes (TC) can also be incorporated by augmenting the model with dummy regressors once the outlier type is identified, following the approach popularised in the X-12-ARIMA literature.

Is Robust SARIMA the same as SARIMA with robust standard errors?

No. Robust standard errors (HAC, sandwich estimators) correct the covariance matrix of coefficient estimates for heteroscedasticity or autocorrelation but do not change how the coefficients themselves are estimated. Robust SARIMA changes the objective function so that outlying observations have less influence on the coefficient estimates themselves.

Sources

  1. Muler, N., Peña, D., & Yohai, V. J. (2009). Robust estimation for ARMA models. The Annals of Statistics, 37(2), 816–840. DOI: 10.1214/07-AOS570 ↗
  2. Franses, P. H., & Ghijsels, H. (1999). Additive outliers, GARCH and forecasting volatility. International Journal of Forecasting, 15(1), 1–9. DOI: 10.1016/S0169-2070(98)00053-3 ↗

How to cite this page

ScholarGate. (2026, June 3). Robust Seasonal Autoregressive Integrated Moving Average Model. ScholarGate. https://scholargate.app/en/econometrics/robust-sarima-model

Related methods

ARIMA modelRobust RegressionSARIMA modelX-13ARIMA-SEATS

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • ARIMA modelEconometrics↔ compare
  • Robust RegressionStatistics↔ compare
  • SARIMA modelEconometrics↔ compare
  • X-13ARIMA-SEATSEconometrics↔ compare
Compare side by side →

Similar methods

Robust ARIMA modelRobust ARMA ModelRobust Time Series AnalysisRobust AR modelRobust MA modelStructural Break SARIMA ModelSARIMA modelBayesian SARIMA Model

Related reference concepts

M-Estimation and Empirical ProcessesModel-Based ClusteringEM AlgorithmJackknife ResamplingMaximum Likelihood EstimationEconometrics

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Robust SARIMA model (Robust Seasonal Autoregressive Integrated Moving Average Model). Retrieved 2026-07-21 from https://scholargate.app/en/econometrics/robust-sarima-model · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
Muler, Peña & Yohai (robust ARMA); earlier foundation by Denby & Martin (1979)
Year
1979–2009
Type
Robust time-series model
DataType
Univariate seasonal time series with possible outliers or heavy-tailed errors
Subfamily
Econometrics / time series
Related methods
ARIMA modelRobust RegressionSARIMA modelX-13ARIMA-SEATS
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account