As of May 31, 2023, we have updated our Code of Conduct.

A vast area which includes generating results from computer models.

Using some model of a "reality", a simulator is run which generates results. A trivial example would be a computer pseudo-random number generator for the Normal distribution. More complex examples include, running simulated cars into barriers to look at their behaviour in impact and numerical weather forecasting where lots of options are run though ensemble models and the most likely is chosen.

Simulation is almost always cheaper and faster than the real thing. Sometimes simulation is the only reasonable ethical solution, e.g. when you are evaluating the safety of a device for human consumption, you can't ethically test it with humans where there is a risk of harm.

In statistics, simulation is often used to test new statistical algorithms - you simulate some data with known parameters, and then you test how well your new algorithm can identify those known parameters and compare your results against older algorithms. In real life you often do not know the true values of the actual parameters.