Year: 2024
Type: conference-paper
Abstract: A common task in economics is the seasonal adjustment of time series, which involves removing the seasonal component from the data. Currently, at the National Statistics Institute (Spain), this task is performed using the Tramo-Seats methodology. Nevertheless, the time series currently being processed extend over many years, which complicates the identification of a single reg-ARIMA model that adequately describes the behaviour of the entire series. New general methodologies are suggested to perform seasonal adjustment on long time series that change its structure due to the effect of a certain event, with two identified models and a transition period. The series before and after the event are considered to be modelable using ARIMA models, while the transition period is modeled as a weighted average of the other two events through a time-dependent weighting function.
Institution: Universidade da Coruña
Language: English
Subfield: Signal Processing
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W4407415204