on 2020 May 15 12:54 AM
When i train my time series model in SAP Analytics Cloud, the model generates a forecast and also identifies the signal outliers in the model.
Q1) Does the MAPE include or exclude the outliers in the calculation
Q2) Is there a way to include / exclude the outliers in the model and see the impact on the MAPE?
Thanks
To add to my previous answer, outliers are being used as part of the MAPE calculation and there is currently no way to include/exclude these values (apart from massaging the initial data source used to train the model). Kind regards Antoine
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hello Ray, when Smart Predict estimates the trend, outliers are actually not being taken into account to not penalize or influence the trend too heavily. Kind regards Antoine
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
67 | |
8 | |
8 | |
6 | |
6 | |
6 | |
6 | |
6 | |
5 | |
5 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.