You can manage the Alpha and Beta values for all your statistical forecast methods on this page.

You can add, edit, or delete Alpha and Beta values for each of the available forecasting methods. These are: Simple Exponential Smoothing, Crostons Method, Modified Crostons, and Double Exponential Smoothing.

The page displays the current values for each method.

Alpha (Base)

Parameters

ParameterValue
Value Display NameThis column displays the names of the values which have been previously created in this page and are available for use within D2000 - Global Stat Parameters.
AlphaYou can edit the effective values in this column. The values entered here will show in the first column as the values which are available for selection when setting parameters in D2000 - Global Stat Parameters.
Delete?Check this box in the Alpha value row to flag it for deletion.
Delete <method> Alpha / BetaSelect this button to delete selected Alpha / Beta values.
Add <method> Alpha / BetaYou can create new values for Alpha or Beta, in any method, with this option.

Steps

Note: The procedure is identical for any of the forecasting methods, and follows the same pattern.

To edit a value for Alpha or Beta, enter a value in the Alpha / Beta column of the desired method. Press Enter to commit. The first column updates to reflect the new effective value.

To add a new value to the Alpha or Beta, for any method:

  1. Select the Add <method> Alpha / Beta button. An add dialog displays.
  2. Select Submit. A new line appears in the method's grid card.
  3. Enter a value for the new Alpha / Beta.

To delete a value, for any of the methods:

  1. Select the Delete? checkbox for the value you want to delete.
  2. Select the Delete <method Alpha / Beta button.

Additional information

All Alpha and Beta values must fall in the range of 0 to 1. 

  • Alpha values control the calculation. A smaller value (closer to 0) creates a smoother (slowly changing) line similar to a moving average with a large number of periods. A high value for alpha tracks the data more closely by giving more weight to recent data. 
  • Beta values control trend calculation. A smaller value considers the longer-term trend and a larger value focuses on the shorter-term trend.

Having many alphas and betas can significantly increase workspace size.