Structural decomposition of time series with implications in economics,accounting, and finance research

Recent literature on the decomposition of a time series into components has produced several alternative methodologies. This situation has arisen because of a lack of consensus on how components should be defined, and on the appropriate decomposition structure. Uncertainty about components structure seems to be inevitable when it is accepted that the behavior of a typical business or economic time series implies evolving rather than fixed components patterns. None of the existing procedures can be viewed as uniquely appropriate for the isolation of unobserved components.