makedist

Create probability distribution object

Syntax

pd = makedist(distname)

pd = makedist(distname,Name,Value)

list = makedist

Description

pd = makedist(distname) creates a probability distribution object for the distribution distname, using the default parameter values.

Use makedist to specify uniform, normal, multinomial, piecewise linear, or triangular distribution objects. If you have Statistics and Machine Learning Toolbox™ software, you can use makedist to create objects for other distributions, such as the Gamma or Weibull distributions. For more information, see makedist (Statistics and Machine Learning Toolbox).

To truncate the probability distribution to a specified interval, use truncate (Statistics and Machine Learning Toolbox).

example

pd = makedist(distname,Name,Value) creates a probability distribution object with one or more distribution parameter values specified by name-value pair arguments.

example

list = makedist returns a cell array list containing a list of the probability distributions that makedist can create.

Examples

collapse all

Create Normal Distribution Object

Open Live Script

Create a normal distribution object using default parameter values.

pd = makedist('Normal')

pd = 
  NormalDistribution

  Normal distribution
       mu = 0
    sigma = 1

Specify Parameters for a Normal Distribution Object

Open Live Script

Create a normal distribution object with a mean value of mu = 75, and a standard deviation of sigma = 10.

pd = makedist('Normal','mu',75,'sigma',10);

Specify Piecewise Linear Distribution Object

Open Live Script

Create a piecewise linear distribution object for a distribution with values from 6 to 10, where values from 6 to 8 are four times more likely than values from 8 to 10.

pd = makedist('PieceWiselinear','x',[6 8 10],'Fx',[0 0.8 1]);

You specify the cumulative distribution function Fx as [0 0.8 1] because the difference between 0.8 and 0 is four times the difference between 1 and 0.8. As a result, the generated distribution has four times more values between 6 to 8 than between 8 to 10.

The plot shows the specified cumulative distribution function (CDF) and the corresponding probability distribution function (PDF).

CDF plot on top and PDF plot on the bottom

The piecewise linear CDF corresponds to a piecewise constant PDF.

Input Arguments

collapse all

`distname` — Distribution name
`'Uniform'` | `'Normal'` | `'Multinomial'` | `'PiecewiseLinear'` | `'Triangular'`

Distribution name, specified as one of the following values:

Distribution Name	Description
`'Uniform'`	Uniform distribution — You specify the lower and upper bounds of the distribution.
`'Normal'`	Normal distribution — You specify the mean and standard deviation of the distribution.
`'Multinomial'`	Multinomial distribution — In a multinomial distribution, the outcome is one of `1`, `2`, ..., `k`. You specify the probability of each outcome, [p1,p2,..., pk]. The probabilities should sum to `1`.
`'PiecewiseLinear'`	Piecewise Linear distribution — Use this distribution to make a custom distribution by specifying a piecewise linear cumulative distribution function (CDF). You specify the vector of values, `x`, at which the CDF changes slope and the corresponding vector of CDF values, `Fx`. The `Fx` value at any `x` is the probability of getting a value less than or equal to `x`. A piecewise linear CDF is created by linearly connecting the known CDF values, `Fx`. This piecewise linear CDF corresponds to a piecewise constant probability distribution function. For an example, see Specify Piecewise Linear Distribution Object.
`'Triangular'`	Triangular distribution — You specify the lower limit, peak location, and upper limit of the distribution.

For information about these distributions, see the Probability Distributions (Statistics and Machine Learning Toolbox) category.

Note

If you have Statistics and Machine Learning Toolbox software, you can use makedist to create objects for other distributions, such as the Gamma or Weibull distributions. For more information, see makedist (Statistics and Machine Learning Toolbox).

Name-Value Arguments

Specify optional pairs of arguments as Name1=Value1,...,NameN=ValueN, where Name is the argument name and Value is the corresponding value. Name-value arguments must appear after other arguments, but the order of the pairs does not matter.

Before R2021a, use commas to separate each name and value, and enclose Name in quotes.

Example: makedist('Normal','mu',10) specifies a normal distribution with parameter mu equal to 10, and parameter sigma equal to the default value of 1.

Multinomial Distribution

collapse all

`Probabilities` — Outcome probabilities
`[0.500 0.500]` (default) | vector of scalar values in the range [0,1]

Outcome probabilities for a multinomial distribution, specified as a vector of scalar values in the range [0,1]. The probabilities sum to 1 and correspond to outcomes [1, 2, ..., k], where k is the number of elements in the probabilities vector. This argument is valid only when distname is 'Multinomial'.

Example: 'Probabilities',[0.1 0.2 0.5 0.2] gives the probabilities that the outcome is 1, 2, 3, or 4, respectively.