This page enables you to display statistical forecast values for both system-optimized and manually-specified values to provide easy comparison of the results, with the double exponential smoothing method.

You can select a product-customer combination from the context selectors at the top right of the page, and display an overview of the respective System Optimization Results for the double exponential smoothing optimization method.

This pane shows the forecast basis input from which the forecast is calculated. Input to stat is set in D2000 - Global Demand Parameters. Where there is more than one forecast basis, for example if the demand forecast basis for customer-product A has been set to Corrected History, and for customer-product B has been set to Pure Demand, then the Mixed Data Check will be highlighted in pink. To make sure an accurate view of statistical forecasts when you select a customer hierarchy level higher than customer-product at leaf level then the forecast basis for all customer-product combinations must be aligned to the same forecast basis.

Current parameter values are shown in the Effective Parameters panel. You can configure the parameters in the Edit Parameters panel. When you update parameters, the new values display in the Effective Parameters panel.

ParametersValues
DES - Optimize

This specifies if the optimization of values is handled by the system:

  • A value of Yes means the system optimizes values itself.
  • A value of No enables the SES - Alpha value to be considered for optimization.
DES - Alpha

Enter a value here to override the global D2000 - Global Stat Parameters setting for the selected customer-product record.

The DES - Optimize parameter must be set to No to make this value effective. If the DES - Optimize value is set to Yes, it is still possible to enter a value here. However, the value will be ignored.

DES - Beta

Enter a value here to override the global D2000 - Global Stat Parameters setting for the selected customer-product record.

The DES - Optimize parameter must be set to No to make this value effective. If the DES - Optimize value is set to Yes, it's still possible to enter a value here. However, the value will be ignored.

DES - Dampen

Enter a value here to override the global D2000 - Global Stat Parameters setting for the selected customer-product record.

The DES - Optimize parameter must be set to No to make this value effective. If the DES - Optimize value is set to Yes, it's still possible to enter a value here. However, the value will be ignored.

DES - Phi

Enter a value here to override the global D2000 - Global Stat Parameters setting for the selected customer-product record.

The DES - Optimize parameter must be set to No to make this value effective. If the DES - Optimize value is set to Yes, it's still possible to enter a value here. However, the value will be ignored.

  • DES – Alpha is the smoothing parameter for the level component. It represents how much weight is given to the most recent observation when calculating the smoothed level.
  • DES – Phi is the smoothing parameter for the trend component of the time series data.
  • DES - Dampen is a smoothing parameter for the graph, and takes on a value between 0 and 1. The damping factor is usually 1 minus the alpha level (1 – α).


This chart displays the system-optimized results for the forecast generated by the double exponential smoothing method. The chart displays this data:

  • Testing Window
  • Effective Adjusted History
  • Optimized Double Exponential Smoothing Forecast

This chart displays the system-optimized results for the best-fit analysis generated by the double exponential smoothing method. The chart displays this data:

  • Testing Window
  • Effective Adjusted History
  • Optimized Double Exponential Smoothing Forecast