AnyLogic 9
Expand
Font size

logistic

The logistic distribution is an unbounded continuous distribution which is symmetrical about its mean [and shift parameter], alpha. The shape of the logistic distribution is very much like the normal distribution, except that the logistic distribution has broader tails.

The logistic function is most often used a growth model: for populations, for weight gain, for business failure, and so on. The logistic distribution can be can be used to test for the suitability of such a model, with transformation to get back to the minimum and maximum values for the logistic function. Occasionally, the logistic function is used in place of the normal function where exceptional cases play a larger role.(see Johnson et. al.1)

Example

beta = 1; alpha = 0

logistic(double beta, double alpha)

Description
Generates a sample of the logistic distribution.
Parameters
Name Type of value Description
beta double The scale parameter > 0.
alpha double The shift parameter.
Result
Type Description
double The generated sample.

logistic(double beta, double alpha, java.util.Random r)

Description
Generates a sample of the logistic distribution using the specified random number generator.
Parameters
Name Type of value Description
beta double The scale parameter > 0.
alpha double The shift parameter.
r java.util.Random The random number generator.
Result
Type Description
double The generated sample.

This document includes content from the “Stat::Fit User’s Manual”. Copyright 2016 Geer Mountain Software Corp.

How can we improve this article?