A modification of time series forecasting methods for handling announced price increases

This paper deals with a modification of exponential smoothing forecasting methods to allow for announced price increases. Such price increases occur frequently and at irregular intervals of time. The authors suggest a procedure that consists basically of an adjustment of sales plus exponential smoothing, moving indices to normalize the original sales data, and modification of the forecast. Numerical results are reported for several time series. The approach is particularly useful for forecasting in an economy with a high inflation rate.