on 01-18-2010 8:44 AM
hi gurus
In which scenerio do we use Error total, Mean squear Error, Squar Root of Mean Error, Mean Percentage Error???
Dear dallyanusha ,
on the following link you can find these Forecast Accuracy Measurements:
http://help.sap.com/saphelp_scm70/helpdata/EN/ac/216b95337b11d398290000e8a49608/frameset.htm
These are used in Univariate Forecasting.
Univariate forecasting allows you to measure the forecast error in six ways:
● Mean absolute deviation (MAD)
● Error total (ET)
● Mean absolute percentage error (MAPE)
● Mean square error (MSE)
● Square root of the mean squared error (RMSE)
● Mean percentage error (MPE)
The system calculates the forecast errors by comparing the differences between the actual values and the ex-post values.
Use the forecast error measurements to help you evaluate the accuracy of the forecast.
You select them in the univariate forecast profile. To set upper limits for the forecast error measurements, use the Diagnostic group in the univariate forecast profile; when a limit has been exceeded, the system issues an alert.
About univariate forecasting you can read more here:
http://help.sap.com/saphelp_scm70/helpdata/EN/47/554b15f8943940e10000000a421138/frameset.htm
I hope I could help.
Regards,
Tibor
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
Hi Dallyanusha,
The scenario of usage of different errors in determining the forecast
accuracy purely depends upon the business decision involved.
It has to be selected based on the type of business data, trend by
which it is failing continuously.
Typically, business analysts who is expert on statistics has to
analyse and review the business data and decide on the scenario to
be followed.
Regards
R. Senthil Mareeswaran.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.
User | Count |
---|---|
8 | |
4 | |
3 | |
2 | |
2 | |
1 | |
1 | |
1 | |
1 | |
1 |
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.