Analysis Section
The fit between the forecast method and historical data is expressed as a Mean Absolute Deviation (MAD) and a Symmetric Mean Absolute Percent Error (SMAPE). In the case of a time-series forecast, this error is the difference between the one-period-forward forecast and historical actuals.
Typically, selecting a method that reduces the error improves your forecast. However, this is not always true. You may find cases where CCH Tagetik Supply Chain Planning selects a method that does not produce the best fit to historical data, but does produce the most reliable forecast.
In the Analysis section, there is a list of several options for viewing data:
Analysis
| Option | Description |
|---|---|
| Residuals Graph | Helps you choose a forecast method as you see the difference between actuals (blue line) and forecast (red line). You can apply different methods to find the best fit. There should be no pattern if the method you have applied is correct. Errors are also displayed.On the right is a histogram of the data. The histogram should look like a standard bell curve. Everything you did not capture is noise. If you see a pattern, you have missed something; perhaps seasonality is not captured. If you see a U shape (showing positive and negative values), this indicates a lifecycle effect is not being picked up. Residual Graph Analysis |
| Year Over Year | Shows the historical data plus the forecast for each year. Year Over Year Graph Analysis |
| Triangle Report | Displays forecast continuity. Since this report depends on checkpoints, it is useful after your system has been deployed for awhile. The bold numbers are actuals while the other numbers are preceding forecasts. Review to see if your forecast has changed dramatically and to pinpoint why. Triangle Report Analysis This is the same report as the Triangle report type, which looks at multiple records. See Work with Triangle Interfaces. |