Benford's Law
 |
|
Description: This example demonstrates how to test probabilities. It uses an unusual example. You might expect that the first digit of laboratory data to be uniformly distributed. Instead it is empirically found that these digits follow an unusual and asymmetric distribution. This phenomenon is known as Benford's Law. It is used some times in cases of scientific fraud for determining the strength of evidence against the given data.
Note the literature references are contained within the script header.
Note the script creates a summary data table with all of the digits (1-9) and their count. They are sorted in ascending order. The Distribution platform displays the first digit information and the significance test of the expected frequencies. A script is saved in the new data table to reproduce the distribution analysis.
Instructions: Open the data table first. Then open the script file and run it. Select the data column that you wish to test.
Script: BenfordsLaw.JSL is the script file
Data: A data table is provided with a set of numbers to test. Feel free to use it with any numeric data column in any data table, though.
SAS Campus Drive, Building S, Cary, NC, 27513 • Phone: 1.919.677.8000© 2008 SAS Institute, Cary, NC, USA. All Rights Reserved. Terms of Use • Privacy Policy