Forecasting > Analysis > Actions > Best Fit Page > Best Fit Management > Best Fit Management Page (Fields)
Best Fit Management Page (Fields)
Forecast
The following describes each field in the section.
Field
Description
Forecast Method
Indicates the algorithms used to calculate the forecast demand.
To run a different Forecast Method:
1. Select the corresponding radio button that is next to the Forecast Method
2. Click
3. The selected Forecast Method will be used the next time the Forecasting process is run. Any subsequent runs will show the newly selected Forest Method as the 2nd row
Months or Slices
The calculated forecast demand for the selected forecast method.
RMSE
Current Root Mean Square Error, which is a statistical measure of fit (only calculated with new period).
MAPE
Current MAPE, only calculated with new period.
* 
In the event of negative demand, the MAPE value can be greater then 2.
MAD
Current Mean Absolute Deviation, which is a statistical measure of fit (only calculated with new period).
Composite Error
Tracking Signal
A measurement that indicates whether the forecast average is keeping pace with any genuine upward or downward changes in demand. It monitors the forecast to see if it is biased high or low. As used in forecasting, tracking signal is the number of mean absolute deviations that the forecast value is above or below the actual occurrence.
Tracking Signal = (RSFE / count) / MAD
• RSFE = Running Sum of Forecast Errors include all slices, demand and archived forecast
• MAD = Mean Absolute Deviation
• Count = Forecast parameter "# of Slices for Forecast Error Calculation" or Best Fit parameter "Holdout Window".
* 
You will only see this field when Enable Tracking Signal is set to Yes for the Forecast Parameter of the selected stream. The Tracking Signal is computed for each SKU stream and the results are displayed as follows:
• On the Demand page
• On the Best Fit Management page
• As review board records for review types 138 and 139
Status
Displays the results of the Best Fit Analysis Rules that were applied to the Forecast Method.
Parameters
The following describes each field in the section.
Field
Description
# of History Slices
The number of demand history periods that the AutoPilot is to use in computing the demand forecast and in calculating the standard deviation in demand factored into Safety Stock.
Winters Multiplicative, Crostons, and Same As Last Year forecast methods require at least one year of history slices. All other forecast methods can use less than one year of history slices.
# of Horizon Slices
The number of periods of forecasting that are projected.
* 
While forecasting 12 months into the future can and does provide valuable planning information, a forecast is best when only viewed a few months out.
Best Fit History Slices
If the maximum number of rolling forecasts is null (default), then:
Best Fit History Slices = Minimum # of History Slices
But, if maximum number of rolling forecasts is not null, then:
Best Fit History Slices = Maximum # of History Slices - Maximum # of Rolling Forecasts - Holdout Window + 1
Holdout Window
The maximum number of history slices to use in comparison to the forecast.
# of Slides
The number of times that MAPE, MAD, and RMSE are calculated using the holdout window positioned within the forecast window. The holdout window is initially aligned with the first time slice in the forecast window, and the error values are calculated. Best Fit will then "slide" the holdout window one time slice later and recalculate the error values until the number of slides is reached. The resulting error values are averaged over the number of slides.
When Double Exponential Smoothing is selected as forecasting method, you will see these additional fields:
Field
Description
Alpha
A smoothing constant used in the smoothing equation for updating Level.
Beta
A smoothing constant used in the smoothing equation for updating Trend.
Phi
A dampening factor for trend estimates. The smaller phi is, the more the trend value will be dampened. If phi = 1, then no trend dampening occurs.
When Moving Average is selected as forecasting method, you will see these additional fields:
Field
Description
# of Slices
The number of history slices used for the forecast calculation.
Weight
The weight factor applied to the forecast calculation. The weight factor takes a value between 0 and 1 inclusive.
When Intermittence Smoothing is selected as forecasting method, you will see this additional field:
Field
Description
Alpha
A smoothing constant used to smooth the level or the base.
When Single Exponential Smoothing is selected as forecasting method, you will see these additional fields:
Field
Description
Alpha
A smoothing constant used in the smoothing equation for updating Level.
When Winters Multiplicative is selected as forecasting method, you will see these additional fields:
Field
Description
Alpha
A smoothing constant used to smooth the level or the base.
Beta
A smoothing constant used to smooth trend estimates.
Gamma
A smoothing constant used to smooth seasonality estimates
Was this helpful?