Ansicht
Dokumentation

GRRMSIMPERCENTV - Maintain Simulation Percentile

General Data in Customer Master   Vendor Master (General Section)  
This documentation is copyright by SAP AG.

In this Customizing activity, you specify the simulation percentile to be used during simulation runs to be carried out for a Monte Carlo simulation in the Risk Management application. The percentile is used to calculate the worst-case scenario for the simulation.

A percentile is the value of a variable below which a certain percent of observations falll. This means that the 20th percentile is the value (or score) below which 20 percent of observations are found.

Note that the simulation percentile is also called the "confidence interval".

The simulation percentile defines the number of impact values to be used during simulation. Impact values refer to the monetary loss of occurring risk events. The impact value can also be regarded as the the potential loss in value of an open item, or the upper loss limit which will not be exceeded over a defined time period, based on a fixed probability.

The formula stored for this and used in the Risk Management application is Certainty = (1 – simulation percentile) * 100. This means that for a simulation percentile of 0.2, the certainty value would be 1 minus 0.2 times 100 = 80.

At the end of the simulation run, the system checks the list of all simulations, based on the percentile defined, and outputs a list of all worst-case scenario data for it. The average value for the percentile is 0.5 (this is the mean between the two values 0 and 1).

1. Click on New Entries and enter a percentage value between 0 and 1.
2. If this is the value to be used in the application for Monte Carlo simulations, checkmark the column Active.
3. Save your entry.

Users define the value at risk as the top 5% (or 1%, or 0.1%, based on the configuration specified in this Customizing activity). This means that:

• If the number of runs in the simulation is 10,000 and the value at risk definition is 5%, the value at risk will be the lowest of the top 500 results of the runs.

• If the number of runs in the simulation is 10,000 and the value at risk definition is 1%, the value at risk will be the lowest of the top 100 results of the runs.

• If the number of runs in the simulation is 10,000 and the value at risk definition is 0.1%, the value at risk will be the lowest of the top 10 results of the runs.

After simulation, the system outputs the results of 10,000 runs for risks with a normal distribution and a Bernoulli frequency distribution. If a 1% value at risk was defined, there will be 100 results in the table. The “worst-case scenario” is therefore the lowest of the 100 worst numbers. The result is always a positive value between 0 and 1.

BAL_S_LOG - Application Log: Log header data   CPI1466 during Backup  
This documentation is copyright by SAP AG.

Length: 3058 Date: 20240523 Time: 195344     sap01-206 ( 64 ms )