摘 要
随着经济全球化和市场竞争加剧，会计舞弊事件频发，给利益相关者带来巨大损失，构建有效的会计舞弊风险预警模型具有重要意义。本研究旨在构建一个科学合理的会计舞弊风险预警模型并进行验证，以期为防范会计舞弊提供理论支持与实践指导。基于现有研究成果，选取财务指标、非财务指标以及公司治理结构等多维度变量作为预警指标体系，运用因子分析法对原始数据进行降维处理，提取出能够综合反映企业状况的关键因子，并采用Logistic回归模型建立预警模型。通过收集沪深两市A股上市公司2010 - 2020年的数据进行实证检验，结果表明该模型具有较高的准确率，在样本外测试中也能较好地识别潜在的会计舞弊风险企业。相较于传统单一维度的预警模型，本研究创新性地融入了非财务指标与公司治理结构因素，使模型更加全面地刻画企业的经营状况与内部管理情况，从而提高预警效果。这不仅丰富了会计舞弊风险预警领域的研究内容，还为企业、投资者、监管机构等提供了更为有效的工具来防范会计舞弊风险。
关键词：会计舞弊风险预警；多维度变量；因子分析；Logistic回归模型；非财务指标与公司治理结构
Abstract
With the intensification of economic globalization and market competition, accounting fraud incidents have become increasingly frequent, causing significant losses to stakeholders. Constructing an effective accounting fraud risk warning model is of great significance. This study aims to develop and validate a scientifically sound and reasonable accounting fraud risk warning model to provide theoretical support and practical guidance for preventing accounting fraud. Based on existing research findings, this study selects multidimensional variables including financial indicators, non-financial indicators, and corporate governance structure as the warning indicator system. Factor analysis is employed to reduce the dimensionality of the raw data, extracting key factors that comprehensively reflect the company's condition. A Logistic regression model is then used to establish the warning model. By collecting data from Shanghai and Shenzhen A-share listed companies between 2010 and 2020 for empirical testing, the results indicate that the model has a high accuracy rate and can effectively identify potential accounting fraud risk enterprises in out-of-sample tests. Compared with traditional single-dimensional warning models, this study innovatively incorporates non-financial indicators and corporate governance structure factors, providing a more comprehensive portrayal of the company’s operating conditions and internal management, thereby enhancing the warning effect. This not only enriches the research content in the field of accounting fraud risk warning but also offers a more effective tool for enterprises, investors, and regulatory authorities to prevent accounting fraud risks.
Keywords：Accounting Fraud Risk Warning；Multi-Dimensional Variables；Factor Analysis；Logistic Regression Model；Non-Financial Indicators And Corporate Governance Structure
目 录
摘 要 I
Abstract II
引 言 1
第一章 会计舞弊风险预警模型的理论基础 2
1.1 会计舞弊的概念与特征 2
1.2 风险预警模型的基本原理 2
1.3 理论框架的构建依据 3
第二章 会计舞弊风险指标体系的构建 5
2.1 指标选择的原则与方法 5
2.2 财务指标的筛选与确定 5
2.3 非财务指标的设计与应用 6
第三章 预警模型的构建与优化 8
3.1 模型构建的技术路线 8
3.2 数据处理与变量设定 8
3.3 模型优化与参数调整 9
第四章 预警模型的验证与应用 11
4.1 样本选取与数据收集 11
4.2 模型有效性检验 11
4.3 实证结果分析与讨论 12
结 论 14
参考文献 15
致 谢 16
