Independence Test

The independence test, indep measures the distribution of the random number generator used in datagen in a matrix. Since the test is unbiased, the row in the matrix is expected to be independent of the column.

The test uses the integer range function to create a row and a column. After a row is created, a column is created for that row. The test counts the number of times each combination is created.

This test is validated by the chi square test. If the test passes, the locations produced by datagen are spread evenly across the matrix. The count in each cell of the matrix should match the expected frequency for that cell.

The test will fail once in twenty trials, on the average. If the test fails, re-run the test. It should pass in the next trial.

Syntax

The syntax for indep is:

      indep size columns rows

Example

      indep 10000 6 6


Table of Contents     Job and Record Level Definitions
Field Level Definitions     Datagen Syntax
Supporting Routines     FAQ     Copying