Random Number Generator

Command:

Math -> Random Number Generator

Random numbers showing a particular distribution are often needed in statistics. DataLab therefore offers a random number generator which can be used to create arbitrarily distributed random numbers. Apart from some important pre-defined distributions (i.e. normal distribution, t-, F-, chi-square-, and Weibull distribution) the random number generator can be set up to generate random numbers according to a user-defined function based on 10 pivot points. In addition, the generator allows to create random numbers by combining two normal distributions.

In order to generate random numbers, you first have to select the type of the distribution of the random numbers by selecting it from the box "Type of Distribution". Currently, the following distributions are available:

• Uniform Distribution
• Normal Distribution
• t-Distribution
• F-Distribution
• Chi-Square-Distribution
• Weibull Distribution
• skewed to the left
• skewed to the right
• leptocurtic
• platycurtic
• bimodal
• user defined

After the selection of the distribution type the corresponding parameters for adjusting the distrbution are displayed. The form of the distribution and the histogram of a sample of 10000 random numbers are shown in graphical form. In addition, the four most important parameters (mean, standard deviation, skewness, and kurtosis) are indicated at the bottom margin of the window.

The range of generated random numbers is controlled by the parameters "lower border" and "upper border". Random numbers beyond this range will not be created (even if the selected distribution allowed random numbers outside this range). Thus, the user should be carefully set the range of the random numbers.

Please be aware of the fact that there are two kinds of distributions: (1) parametric distributions (normal distribution, t-, F-, chi-square-, and Weibull distribution) which are controlled by their parameters, and (2) non-parametric distributions which always span the distribution window, no matter how the borders are set.

After setting up the required distribution, you have to click the button "Generate Random Numbers" in order to created the random numbers. The random numbers can fill one or several columns of the data matrix; individual rows cannot be filled.

You may define an arbitrary distribution function by selecting "user defined" in the "Type of Distribution" box. The user-defined distribution function can be specified by setting 10 pivot points (just click and move them with the mouse). The connecting line between the pivot points will form the distribution function. The definition of the pivot points can be saved on disk by clicking the save button at the top right of the window. Please note that saving the pivot points is only available in the full version of DataLab.