Stanojević, Jelena Radojičić, Dragana application/pdf 1236943 bytes info:eu-repo/semantics/conferenceProceedings Keywords: Fraud detection, Benford’s law, statistical tests, resampling methods 51. Симпозијум о операционим истраживањима number: 51 startpage: 545 endpage: 550 In this paper, we use Benford’s law as the main methodology for fraud detection in the financial reports of three private hospitals operating in Serbia. This law is a statistical method that can be used to quickly and efficiently detect suspicious positions and figures in large datasets. The law states that the leading digits from 1 to 9 appear in a decreasing logarithmic distribution. It states that the digit 1 appears most frequently with 30% of the time and the digit 9 has the lowest frequency with about 5% of the time according to this law. We also investigate the conformity of an empirical distribution of observed real data with the Benford’s distribution. The main goal of the paper is to present an application of the permutation tests as well as the bootstrap tests for checking conformity with the law, and the application of resampling methods is proposed. FRAUD DETECTION IN FINANCIAL REPORTS OF THREE PRIVATE HOSPITALS OPERATING IN SERBIA, TESTING WITH THE NEW TESTS http://creativecommons.org/licenses/by-nc-nd/4.0/legalcode https://phaidrabg.bg.ac.rs/o:35554 doi:10.24867/SYMOPIS-2024-51-087 ISBN: 978-86-6022-703-6 Fakultet tehničkih nauka 2024 eng