Remove outliers from historical transaction data when calculating a demand forecast

This article describes how to exclude outliers from the historical data that is used to calculate a demand forecast. By excluding outliers, you can improve forecast accuracy.

You can exclude outliers to improve forecast accuracy. This is an optional task. Here is an overview of the process:

  1. Click Master planning > Setup > Demand forecasting > Outlier removal to open the Outlier removal page, where you can use a query to select the transactions to exclude.
  2. Select the company that the query applies to, and then enter a name and description. The Query date field is automatically set to the current date.
  3. Select the Active check box to exclude the transactions that the query finds from the historical data. This setting will take effect when you create a baseline forecast.
  4. On the Outlier removal query page, you can add, remove, and select the criteria that define which transactions will be excluded when the baseline forecast is calculated. For example, select a specific item or order transaction to exclude.
  5. Click Display transactions. The Outlier transactions page lists the transactions that meet the criteria that you defined in the query, and that will be excluded from the historical data when the demand forecast is calculated.

Note: You can also create a query that is based on an existing query. Select the query to copy, and then click Duplicate. The Query date field identifies the version. You can use the query as it is, or you can click Edit query to modify the criteria. You can optionally modify the name and description of the new query.

See also

Introduction to demand forecasting

Monitoring forecast accuracy