A Comparative Study of Seasonal Adjustment Methods for Philippine Time Series Data (2002)

 

Abstract

This study explored the applications of X12 RegArima and TRAMO-SEATS deseasonalization approaches to some Philippine time series data using a new freeware from EUROSTAT called DEMETRA, which interfaced these two seasonal adjustment methods. Theoretically , explicit model based procedures, such as TRAMO SEATS, ought to be better than the X-11/X-12 family since the latter’s filters do not take into account the statistical properties of the time series while the TRAMO-SEATS filters are mean-squared error optimal. This research undertaking sought to consider, on the grounds of some empirical criteria, which procedure ought to be preferred for some Philippine time series. For the domain of Philippine time series studies, TRAMO-SEATS is recommended.