ABSTRACT
The initial software in seasonal adjustment was developed by the US Census Bureau in 1954, and was based on the moving average method. Most economic data are subject to seasonal variation. Seasonal adjustment produces data in which the values of neighboring months are usually easier to compare. The use of the term seasonality emanates from the fact that seasons exert an unquestionable influence on economic activity. Trading day effects can be estimated by assigning weights to days of the week reflecting the relative level of economic activity. Extreme values, or outliers in time series can affect the trend depending on the extent of this activity, such that it is both necessary to identify and correct these irregular values. It is well known that estimation of seasonal factors, both current and forecast, has a significant role, when seasonality is moving rapidly in a stochastic manner.
