Controlling When a Stochastic Element is
Sampled
By default, Stochastic elements are sampled once per realization A single model run within a Monte Carlo simulation. It represents one possible path the system could follow through time. (typically at the beginning). Hence, if you specify the parameters describing a Stochastic as a function of time, unlike other GoldSim elements, Stochastic elements are not updated (resampled) if the parameters change.
GoldSim does provide a mechanism, however, to force a Stochastic to either 1) be sampled at a specified time; or 2) be resampled multiple times during a simulation. In particular, you can define one or more triggers for the Stochastic which specify when (or under what circumstances) the Stochastic is resampled.
How a Stochastic is resampled is controlled by the Mode specified on the dialog:
The default is for the Stochastic to be Sampled once. However, if Resampled is selected, the Resample... button in the dialog becomes available, and this allows you to force the Stochastic to be resampled (triggered) at other points in the simulation.
Note: You cannot specify triggers to resample a Stochastic if it is correlated to a second Stochastic. However, if a Stochastic is correlated to a second Stochastic, and that second Stochastic is resampled, the correlated Stochastic will also automatically be resampled (even though no resampling triggers are specifically defined, and the Mode is set to Sampled once).
Conditional Containers can also be used to control when Stochastic elements are resampled.
An example file which illustrates how a Stochastic is sampled (BasicStochastic.gsm) is in the General Examples/Stochastic folder in your GoldSim directory (accessed by selecting File | Open Example... from the main menu).
- Applying Importance Sampling to a Stochastic Element
- Browser View of a Stochastic Element
- Controlling When a Stochastic Element is Sampled
- Correlating Stochastic Elements
- Creating a Stochastic Vector
- Customized Importance Sampling Using User-Defined Realization Weights
- Dynamically Revising Distributions Using Simulated Bayesian Updating
- Specialized Functions That Operate on Distributions
- Specifying a Deterministic Value for a Stochastic
- Specifying the Distribution for a Stochastic Element
- Stochastic Distribution Types
- Triggering a Stochastic