SIMSYSTEM Procedure

Simulation is a powerful method for understanding and predicting the behavior of a complex system under different conditions when it is not feasible to experiment with the system and measure how the output changes in response to variation in the inputs. See Ross (2006) for an introduction to the design and analysis of simulation studies.

Probabilistic simulation—often referred to as Monte Carlo simulation—is based on a probability model that assumes distributions for the input variables and assumes a functional relationship between the input variables and the output. By sampling a large number of values from the input distributions, you can simulate the distribution of the output and compute probabilities that quantify uncertainty in the performance of the system.

Simulation based on probability distributions is essential for assessing and managing risk in many fields of business, government, and research where actual data for unusual events are limited or unavailable. These fields include banking (market and credit risk), insurance (actuarial risk), fraud prevention, cybersecurity risk assessment, reliability engineering, national defense, climate modeling, environmental analysis, and epidemiology. In the field of statistics, simulation is a standard approach for evaluating the performance and robustness of statistical methods when applied to data that cannot be represented by standard modeling distributions such as the normal distribution.