Estimates of a seasonal index in the standard manner (from a moving average)introduce systematic error into the seasonal estimates if a trend is present. Thispaper shows that a logarithmic modification of the standard moving average procedurewill cause it to be consistent with a trend and is an efficient alternative. This paperalso compares several other efficient seasonal indexing procedures appropriate for routinebusiness applications and shows some numerical results. The results indicate thatit is possible to achieve an improvement in the precision of the seasonal index, in theseasonally adjusted data and in forecasts based upon this data, by considering logarithmicalternatives to standard seasonal indexing procedures. This improvement may beaccomplished without a substantial increase in complexity or in the associated computationalburden. The opportunities for improvement are shown to be greatest when thedata contain substantial trend and seasonal aspects and when the trend has a percentageform. Some suggestions for forecasters are offered.
展开▼
