The ratio-to-moving-average method of seasonal adjustment was developed during the 1920's by Frederick R. Macauley at the U.S. National Bureau of Economic Research. Ever since, the use of moving averages has remained a standard in procedures to separate seasonal fluctuations from trend/cycle and irregular fluctuations. Based on the premise that seasonal fluctuations over a full cycle (usually a year) add up to zero, it is taken that these fluctuations are canceled out in a moving average with a length of one or several full years. The moving averages can then be compared to the full series as a first step to find an average seasonal pattern.
For some practical purposes, such as company management decisions, payments on inflation-indexed contracts, monetary policy based on current macro-economic developments, a set back of such procedures is that a final result is only reached after some time has elapsed. At the moment when data have news value, only a provisional seasonal adjustment is available.
In this paper it is demonstrated that an arithmetic progression will do about the same job as a centered moving average, but without the need to wait for the centered moving average to reach the actual term. Next it will be looked into whether a sequential procedure of direct, final seasonal adjustment is feasible. One such procedure will be proposed as an alternative to the provisional outcomes of a centered moving average procedure. It will be argued, that the results of the proposed procedure will be closer to the final outcomes of a centered moving average procedure than its provisional results can be.