Review forecast accuracy

A new home for the Coupa Supply Chain documentation

Starting with Supply Chain 42, our documentation will be located on Coupa Compass. The help here will continue to be accessible for the foreseeable future, but will no longer be updated. If you need a Compass user account, contact your company’s Designated Support Contact (DSC).

You can access new and updated Supply Chain documentation in the following location on Compass:

https://compass.coupa.com/en-us/products/supply-chain-design-and-planning

Review forecast accuracy

Log into the Demand Modeler app using the link provided by your administrator.
Select the Output Overview tab.
Select the Forecast Accuracy button. The Forecast Accuracy page appears.
After completion of the forecast run, use the Forecast Accuracy page to continue to make input parameter changes around the models, causals and demand history treatment to further tweak and improve your forecast accuracy. This tab also includes the Top 10 products by demand, showing which products account for the majority of your demand.
Before you view your forecasts, be sure you have run the appropriate scenarios.
Select a scenario to use from the Scenario field at the top of the page.
Select the Filters button to select the slice of the forecast to analyze. The visualizations are updated according to your selections.
Review the results.

Forecast Accuracy tab

Visualization	Description
Demand	The total demand for the product(s) selected by the filters.
Time Series
wMAPE	The weighted mean absolute percentage error (wMAPE) for the products selected. In the context of demand forecasting, weighted mean absolute percentage error (WMAPE) is equivalent to WAPE when the errors are weighted by sales volume. The WAPE is calculated by weighting the error between the actual values and forecasts with the sum of the actual values over all units of time. A lower value of MAPE implies a higher accuracy of the model.
RMSE	The root mean square error (RMSE) for the selected products. The RMSE is the square root of the average squared errors between the actual values and forecasts for each unit of time. This metric is an extension of the mean squared error. A lower value of RMSE implies a higher accuracy of the model.
MAE	The mean absolute error (MAE) for the selected products. The MAE is the mean of absolute differences between actual values and forecasts for each unit of time. A lower value of MAE implies a higher accuracy of the model.
Bias%	The percent bias for the selected products. The Bias% measures the average tendency of the modeled values to be larger or smaller than their observed values. A lower value of Bias% implies a higher accuracy of the model.
Top 10 Products (wMAPE)	This graph displays the products with the lowest wMAPE results for the selected products. These are the products with the most accurate forecasts.
wMAPE Distribution	This graph displays the wMAPE results by the number of products for the selected products.
Demand, Forecast, & Causals	This graph displays the historical and forecasted change in demand over the selected time period for the selected products. The left vertical axis shows demand quantity, and the right vertical axis shows the effect of the causal. Select any causals to add to the graph from the Causals field.
Model Selection	This graph displays the number of products for each model type for the selected products.
Error Metrics	This table displays the model used for each dimension combination for the selected products and the resulting error metrics for that forecast.

Last modified: Friday May 12, 2023

Is this useful?

Yes No