**Select menu: Data | Generate Random Sample**

This lets you generate a pseudo-random sample from a selected distribution. You can then specify the size of the sample and the parameters of the distribution from which it is drawn. The distribution can be specified in terms of its mean and variance, or in some cases by the parameters of the distribution.

- After you have
**imported your data, from the menu select Data | Generate Random Sample**.

## Distribution

Specifies the distribution from which the sample is to be drawn.

- Beta
- chi-square
- exponential
- F
- gamma
- log-normal
- normal
- t
- uniform
- Weibull
- binomial
- hypergeometric
- Poisson

## Size of sample

Sets the size of sample to be generated.

## Seed

Sets the seed for the random number generator. A value of 0 will use a random seed generated from the system clock if this is the first use of the generator; otherwise it will continue the sequence previously used. If the seed is taken from the system clock a message will be produced in the output indicating its value, allowing it to be re-used.

Any particular non-zero integer seed will always produce the same underlying sequence of pseudo-random numbers before transformation to the required distribution, so the seed should normally be set to 0 except when you wish to exactly repeat the random sample.

## Save in

Specifies the name of a variate to save the sample in. You can double-click an entry in the variable list or type the name directly into the field.

## Specify parameters

Selecting this option lets you specify the selected distribution in terms of its parameters.

## Properties

Lets you define the distribution according to various moments or parameters depending on the setting of **Specify parameters**. See the documentation on each distribution for a description of how it may be defined.

## Display in spreadsheet

Lets you display the results in a spreadsheet. You can select the sheet from the list of current open spreadsheets or request a new spreadsheet be created. Note: the number of rows of the spreadsheet must match the length of the results formed by the calculation, otherwise a new sheet will be used.

## See also

- GRANDOM procedure which is used to generate the random sample