Replication Data for: Fraud Triangle Perspective: Artificial Neural Network Used in Fraud Analysis

Abstract This study aims to determine the Triangle Perspective: Neural Network Used in Fraud Analysis. Application of the Artificial Neural Network method to investigate and detect fraud in companies. The number of samples used in this study was 102 companies. The population in this study are manufacturing companies listed on the Indonesia Stock Exchange for the 2016-2019 period. The test was carried out using an artificial neural network (ANN) method using the SPSS 25 application. External pressure does not affect fraudulent financial statements, and financial stability significantly affects financial statement fraud. The nature of the industry has a significant effect on financial statement fraud, inventory to sales ratio has a significant effect on financial statement fraud, the ineffective monitoring variable has no significant effect on financial statement fraud, gross profit to total assets has a significant effect on financial statement fraud. Return on assets has no significant effect on financial statement fraud, net profit margin (NPM) has no significant effect on financial statement fraud, firm liquidity (WCTA) has a significant effect on financial statement fraud. Keywords: artificial neural network, financial statement fraud, pressure, opportunity, rationalization

摘要 本研究旨在探究三角视角下应用神经网络开展欺诈分析的相关议题。本研究采用人工神经网络（Artificial Neural Network, ANN）方法，对企业欺诈行为开展调查与侦测。本研究共选取102家企业作为研究样本，研究总体为2016-2019年在印度尼西亚证券交易所上市的制造业企业。本研究借助SPSS 25统计软件，采用人工神经网络方法开展检验。
研究结果表明：外部压力对财务报表欺诈无显著影响；财务稳定性对财务报表欺诈具有显著正向影响；行业性质对财务报表欺诈具有显著影响；存货销售比对财务报表欺诈具有显著影响；监督无效变量对财务报表欺诈无显著影响；毛利润与总资产之比对财务报表欺诈具有显著影响；资产收益率（Return on Assets, ROA）对财务报表欺诈无显著影响；净利润率（Net Profit Margin, NPM）对财务报表欺诈无显著影响；企业流动性指标营运资本与总资产之比（Working Capital to Total Assets, WCTA）对财务报表欺诈具有显著影响。
关键词：人工神经网络（Artificial Neural Network, ANN）、财务报表欺诈、压力、机会、合理化