Conformity of census data
Before using Benford's law, it's important to check whether your dataset conforms to Benford's law. The dataset census.2009
contains the populations of 19509 towns and cities of the United States (July 2009) and was used in Nigrini, M. J. (2012). Benford’s Law: Application for Forensic Accounting, Auditing and Fraud Detection. Wiley and Sons: New Jersey. Let's test if the populations in 2009 satisfy Benford's law.
The plots will show the graphs and the legends separately, click Previous Plot to check the graphs.
This exercise is part of the course
Fraud Detection in R
Exercise instructions
- Load the
benford.analysis
package. - Check the conformity of the variable
pop.2009
to Benford's Law distribution of the first digit. Note that we only look at the first plot that is produced. The other plots are outside the scope of this course. - Multiply the variable
pop.2009
by 3 and check the conformity again.
Hands-on interactive exercise
Have a go at this exercise by completing this sample code.
# Load package benford.analysis
library(___)
data(census.2009)
# Check conformity
bfd.cen <- benford(___$___, number.of.digits = ___)
plot(bfd.cen, except = c("second order", "summation", "mantissa", "chi squared","abs diff", "ex summation", "Legend"), multiple = F)
# Multiply the data by 3 and check conformity again
data <- census.2009$pop.2009 * ___
bfd.cen3 <- benford(___, number.of.digits=___)
plot(bfd.cen3, except = c("second order", "summation", "mantissa", "chi squared","abs diff", "ex summation", "Legend"), multiple = F)