To learn more about EpiX Analytics' work, please visit our modeling applications, white papers, and training schedule.

Page tree

 

Integer Uniform(min,max):

 = DiscreteUniform(min,max), which is a special case of the more general Discrete Uniform distribution that one can construct with Crystal Ball's Custom Distribution

 

Integer uniform equations

 

 

The Integer Uniform distribution is a distribution used to describe a variable that can take one of several consecutive explicit integer values. For example, the resulting value from throwing a die 1 time. An example of the Integer Uniform distribution is shown below:

 

 

Uses

It is not often that we come across a variable that can take one of several equally-spaced values each with equal probability. However, there are a couple of modeling techniques that require that capability:

Bootstrap

Resampling in univariate non-parametric Bootstrap

Fitting empirical distribution to data

Creating an empirical distribution directly from a data set, i.e. where we believe that the list of data values is a good representation of the randomness of the variable.

Bayesian analysis by simulation

An Integer Uniform distribution can represent an uninformed prior for a discrete variable, so it is useful when you perform a Bayesian inference analysis with simulation using accept/reject criteria. The pats model is an example.

 

Generation

The Integer Uniform distribution is not directly available in Crystal Ball 5.5- as a distribution but can be readily created using Crystal Ball's Uniform distribution and Excel's ROUND function as follows:

A1 = 1     (min)

A2 = 5     (max)

B1 = A1 - 0.5

B2 = A2 + 0.5

A3 = Uniform(B1,B2)

A4 = ROUND(A3, 0)

In Crystal Ball 7.0+, you can use the DiscreteUniform Distribution available in the distribution gallery.

 

 

 


  • No labels