Tools › Finance and Business › Auditing › Benford's Law Test › Example

Benford's law, second digit: round and repeated amounts in 903 expense claims

ToolBenford's Law Test
Sourceoriginal
LevelProfessional

Problem

A list of 903 expense amounts between 10.00 and 12,878.71 holds many rounded figures and two amounts that recur. Test the second digit against Benford's law at the 5 % level, list the most repeated amounts and find the share of round amounts.

Inputs to Enter

  • Amounts: 903 seeded illustrative amounts (one per line); smallest amount 10
  • Test in full: second digit; significance level 5 %; no approval limit
Open the Benford's Law Test tool

Answers

  • Benford proportion of the second digit 0 = sum over k = 1 to 9 of log10(1 + 1/(10k)) = 0.1197; expected count 108.1; observed 184 (20.4 %); z = 7.73
  • Second digit 5: 146 amounts where 87.3 are expected, z = 6.55
  • Second digit: chi-square 137.19 on 9 degrees of freedom, p < 0.0001; mean absolute deviation 0.03257; 7 of 10 digits beyond ±1.96 (0, 5, 1, 7, 8, 6, 2)
  • First digit: chi-square 10.95 on 8 degrees of freedom, p = 0.2048: the first digits alone show little
  • Most repeated: 249.99 and 100.00 occur 28 times each, 150.00 22 times, 250.00 20 times; 199 of the 903 entries repeat an amount already in the list
  • Round amounts: 207 whole amounts (22.9 %), 187 multiples of 10 (20.7 %), 80 multiples of 100 (8.9 %)
  • The middle 90 % of the amounts cover 1.66 orders of magnitude, so the page marks the law as an approximation for this list

Notes

Open this case in the tool: /analysis-tools/business/audit/benfords-law-test/?try=round-and-repeated. Seeded illustrative data; checked against an independent numpy and scipy computation. Quoted prices, estimates and fixed fees are round by nature: a deviation is a reason to look at those entries, not evidence of fraud or error.

More Examples for This Tool