Probability distribution functions represent an uncertain quantity. These return results that differ depending on whether they are evaluated in Mid-mode or Sample-mode. When evaluated in Mid-mode, a distribution returns its theoretical median. When evaluated in sample mode, a distribution returns a random sample of points drawn from the indicated probability distribution, where each sample point is indexed by the system index Run. Random samples may be generated using Median Latin Hypercube Sampling, Random Latin Hypercube Sampling or Monte Carlo Sampling, based on the current Uncertainty Settings.
Distribution functions can also be specified as a parameter to the Random function to generate single random variates.
When unexpected dimensions appear in the parameters to a distribution function, the function is array-abstracted across those dimensions, so that independent samples are generated for each element combination of the extra dimensions.
To generate independent samples across a dimension, either (artificially) introduce the function's parameters, or use the Over parameter. For example, to generate independent standard normal samples for each combination of elements in I and J, use:
This category has the following 11 subcategories, out of 11 total.
Pages in category "Distribution Functions"
The following 54 pages are in this category, out of 54 total.