Application of Newcomb-Benford law to aid fraud detection

• Autor(a) principal: Leoni, Roberto Campos
• Data de Publicação: 2020
• Outros Autores: Santos, Graciele Silva Dos, Mendonça, Marialda Mathias, Sampaio, Nilo Antonio Souza, Silva, José Wilson de Jesus
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Research, Society and Development
• Texto Completo: https://rsdjournal.org/index.php/rsd/article/view/5276
• topic: Newcomb-Benford; Accounting model; Z-test; x2 -Test.; Newcomb-Benford; Modelo contabilométrico; Teste-Z; Teste-x2.; Newcomb-Benford; Modelo contable; Prueba Z; Prueba -x2.

The Newcomb-Benford law states that the expected frequencies for the first digits of values that make up certain data sets do not follow a uniform distribution. Satisfied the assumptions of the Newcomb-Benford law, the probability of occurrence of digits 1 to 9 is decreasing, being equal to 30.1% for digit 1, 17.6% for digit 2 and only 4.6% for digit 9, for example. In this article, the Newcomb-Benford law is presented as a possible tool to be used in the field of accounting auditing. The accounting model used is based on relationship between the Newcomb-Benford law and the statistical hypothesis tests: Test-Z and x2-Test. In a practical application, 589 values of commitment grades from the Military Academy of Agulhas Negras were analyzed from January 1, 2017 to June 30, 2017. It was concluded that the data presented conformity with the proposed model in relation to leading digits of the analyzed values and that the accounting model can be used by auditors to detect accounting deviations.