A weighting transformation is commonly used to de-seasonalize your demand using externally supplied seasonal weights, or to normalize the data for trading day effects (e.g., 4-4-5 calendars, number of working days per month, etc.).
To use a weighting transformation, you must create a helper variable containing the weights. Helper variables are not forecasted, and their values are not included in group totals. Helper variables are used in conjunction with event models, by analogy models and weighting transformations.
To apply a weight transformation, select Weights on the More icon drop-down on the Forecasting tab or select Forecasting > Weights > Select on the Navigator’s context menu to open the dialog box shown below.
Select the helper variable with the weights you want to apply and click OK to apply the forecast modifier.
The following modifier is used to specify the weighting transformation:
- \WGT=X: Use a weighting transformation. X is the name of the helper variable time series containing the weights.
The procedure divides each value of the specified time series by the corresponding value (weight) in helper variable X. It then forecasts the de-weighted variable and multiplies the forecasts by their corresponding weights. The weighting variable X must span the entire history and forecast period for each variable to be forecasted.