财务信息是企业各利益相关者进行分析与决策的基础， 高质量的财务信息 不仅可以准确反映企业的财务状况和经营风险，还可以帮助信息使用者做出科 学合理的经济决策。西北五省作为国家“丝绸之路经济带”的重要环节，对于 国家政策的施行有较大影响。但目前关于财务信息质量的评价多以我国上市公 司为研究对象，以西北上市公司为研究对象的较少。因此，对西北上市公司的 财务信息质量进行深入研究，发现其经营过程中的财务信息短板并及时弥补， 从而提高企业的财务信息质量，对西北地区经济的发展十分必要。 首先，论文在梳理财务信息质量特征、研究内容等相关文献的基础上，界 定了财务信息和财务信息质量的内涵，得出财务信息是会计信息的延伸和拓展 的结论。其次，通过分析西北上市公司财务信息质量现状，针对其披露质量较 差的短板，以结果导向为主，过程导向为辅的评价思路，构建了包含3个一级 指标，21个二级指标的西北上市公司财务信息质量评价指标体系。再次，对人 工神经网络结构设计，确定评价模型中的隐含层层数、各层节点数，运用层次 分析法，综合考虑审计意见、财务风险、披露质量作为人工神经网络模型的期 望值。然后将西北上市公司数据标准化处理后输入人工神经网络，经过训练和 检验，得到各指标权重，确定评价模型。最后，应用该模型对西北上市公司进 行评价，结果与西北上市公司的财务信息质量实际情况相符，从而验证了模型 的有效性，并得到以下结论：（1）基于人工神经网络的定量评价模型适用于西 北上市公司财务信息质量的评价。（2）西北上市公司财务信息质量较差的主要 原因是盈利能力较差，现金流不足以及内部控制有缺陷。 关键词：财务信息，质量评价，西北上市公司，人工神经网络 II RESEARCH ON EVALUATION OF FINANCIAL INFORMATION QUALITY OF NORTHWESTLISTED COMPANIES BASED ON ARTIFICIAL NEURAL NETWORK Abstract Financial information is the basis for analysis and decision-making by various stakeholders of the enterprise. High-quality financial information can not only accurately reflect the financial status and operating risks of the enterprise, but also help information users to make scientific and reasonable economic decisions. The five provinces of the Northwest, as important links in the country's "Silk Road Economic Belt," have a greater impact on the implementation of national policies. However, most of the current evaluations of the quality of financial information are based on China's listed companies as research objects, and less on Northwest listed companies. Therefore, an in-depth study of the quality of financial information of listed companies in Northwest China has found that the shortcomings of financial information in the course of their operations and make up for them in a timely manner, thereby further improving the quality of financial information of enterprises, is very necessary for the economic development of Northwest China. First of all, the paper defines the connotation of financial information and the quality of financial information on the basis of combing the relevant characteristics of financial information quality, research content, evaluation methods and evaluation index system, and concludes that financial information is the extension and expansion of accounting information. Secondly, by analyzing the current status of financial information quality evaluation of listed companies in Northwest China, aiming at the shortcomings of poor disclosure quality, focusing on results-oriented and process-oriented evaluation ideas, a three-level index and 21 second-level indicators were constructed. Index system for evaluating financial information quality of Northwest listed companies based on indicators. Thirdly, for the design of the artificial neural network structure, determine the number of hidden layers and the III number of nodes in each layer in the evaluation model, and use the analytic hierarchy process to comprehensively consider audit opinions, financial risks, and disclosure quality as the expected value of the artificial neural network model. Then the data of Northwest listed companies are standardized and input into artificial neural network. After training and testing, the weight of each index is obtained and the evaluation model is determined. Finally, the model was used to evaluate Northwest listed companies, and the results were consistent with the actual financial information quality of Northwest listed companies. The validity of the model was verified and the following conclusions were obtained: (1) The quantitative evaluation model based on artificial neural network is suitable for Evaluation of Financial Information Quality of Northwest Listed Companies. (2) The main reasons for the poor quality of financial information of Northwest listed companies are poor profitability, insufficient cash flow, and flawed internal control.. Keywords：Northwest listed company，financial information，quality evaluation， artificial neural network IV 目 录 中文摘要 ................................................. I Abstract ................................................ II 第一章 绪论 .............................................. 1 1.1 研究背景与意义 .............................................. 1 1.1.1 研究背景 ............................................... 1 1.1.2 研究意义 ............................................... 3 1.2 国内外研究现状 .............................................. 3 1.2.1 财务信息质量特征 ....................................... 3 1.2.2 财务信息质量评价内容 ................................... 5 1.2.3 财务信息质量评价方法与模型 ............................. 6 1.2.4 财务信息质量评价指标体系 ............................... 7 1.3 研究内容与研究方法 .......................................... 8 1.3.1 研究内容 ............................................... 8 1.3.2 研究方法 .............................................. 10 1.4 研究重点与研究难点 ......................................... 11 1.4.1 研究重点 .............................................. 11 1.4.2 研究难点 .............................................. 11 1.5 创新点 ..................................................... 11 第二章 理论基础 ................................................. 12 2.1 会计信息与财务信息 ......................................... 12 2.2 财务信息的分类 ............................................. 12 2.2.1 按表达方式分类 ........................................ 12 2.2.2 按时态分类 ............................................ 13 2.3 财务信息质量 ............................................... 13 V 2.4 人工神经网络概述 ........................................... 14 2.5 相关理论 ................................................... 15 2.5.1 信息不对称理论 ........................................ 15 2.5.2 委托代理理论 .......................................... 16 2.5.3 利益相关者理论 ........................................ 16 第三章 西北上市公司财务信息质量评价指标体系构建 ............ 17 3.1 西北上市公司财务信息质量现状 ............................... 17 3.1.1 西北上市公司财务报告违规情况 .......................... 17 3.1.2 西北上市公司财务信息披露环境 .......................... 17 3.1.3 西北上市公司财务信息披露质量 .......................... 18 3.1.4 西北上市公司财务信息质量对比分析 ...................... 19 3.2 构建原则 ................................................... 19 3.2.1 可操作性 .............................................. 19 3.2.2 系统性 ................................................ 20 3.2.3 重要性 ................................................ 20 3.2.4 成本—效益性 .......................................... 21 3.2.5 科学性 ................................................ 21 3.3 财务信息质量评价标准 ....................................... 21 3.4 财务信息质量评价指标设计 ................................... 23 3.4.1 相关性指标 ............................................ 23 3.4.2 可靠性指标 ............................................ 25 3.4.3 透明度指标 ............................................ 26 第四章 人工神经网络模型设计 ................................... 29 4.1 人工神经网络的结构设计 ..................................... 29 4.1.1 层数的设计 ............................................ 29 4.1.2节点数的设计 .......................................... 29 VI 4.2 人工神经网络的期望值 .