Several seasonal adjustment strategies in problematic contexts

The past few years have been marked by the occurrence of many unexpected events that have had many social and economic repercussions, with the COVID-19 pandemic and rising tensions in energy commodity markets standing out above the others. This period of great uncertainty has also had a considerable effect on the production of official economic statistics, undermining the goodness and the predictive capacity of short-term stochastic models. In this condition of extreme unpredictability, there is a need for a strategy of monitoring and reviewing the seasonal adjustment models and anomalous observations, especially over the period 2020-2023. In this work several intervention strategies were defined and tested, focusing over series that manifested a distinct break in their dynamic. Temporary level shifts, included with their lagged versions, have proven to be a particularly useful tool. The outcomes reveal that the policies we considered are effective, and the TRAMO-SEATS procedure manages to be helpful in both ordinary and extraordinary conditions. The whole data analysis has been conducted with JDemetra+ that is a complete and flexible tool in performing several statistical estimates and tests.