SEATSX12ARIMA

SEATSX12ARIMA is a term used to describe a model-based seasonal adjustment framework that combines SEATS (Signal Extraction in ARIMA Time Series) with the X-12-ARIMA family of methods. It is used to decompose a univariate time series into seasonal, trend-cycle, and irregular components under an ARIMA modelling premise, producing a seasonally adjusted series among the outputs.

The approach fits an ARIMA model to the data and uses a state-space representation to extract signals corresponding to the different components. The seasonal component is estimated and removed to obtain the seasonally adjusted series, while the trend-cycle and irregular components are retained for analysis. The framework can incorporate calendar effects and robustness features to outliers, and it provides diagnostic outputs such as residuals and component series.

SEATSX12ARIMA is part of the broader X-13ARIMA-SEATS family used by statistical offices and researchers for official statistics and economic analysis. The integration aims to combine the strengths of a model-based signal extraction with established seasonal adjustment practices, offering revision-friendly adjustments and cross-vintage comparability. Software implementations expose this backend alongside other SEATS/X-12 options.

Limitations include sensitivity to the specification of the ARIMA model and the quality of the input data. Users should evaluate model diagnostics and consider alternative specifications to ensure robust seasonal adjustments, particularly for series with structural breaks or irregular calendar effects.