基于审计数据的仿真数据源分析及应用研究

发布时间：2018-10-12 10:53

【摘要】：为确保我国经济快速稳定的发展,审计机关应充分发挥其职能,这就要求审计机关要善于预测、识别、化解经济风险,建立对审计风险意识的预测,从错综复杂的问题中找出对经济发展产生主要影响的因素与问题,并将反馈信息及时报告给国家有关部门,提出政策性审计的建议。这些都是基于审计数据展开的,但是,目前的现状是对审计数据的提取需要耗费大量人力、物力、时间,得到的数据有的还具有一定的保密性、异构性,并不能被审计人员很好的利用,这使得有些审计业务很难开展,为此,需要构建一个仿真数据库。本文的研究内容主要针对社会保险领域中的养老保险,基于养老保险行业国家标准库中的9张表153字段分析出养老保险行业表之间、字段之间的相互关系,并通过对真实数据的处理、拟合,利用蒙特卡罗方法生成6张数据表中50个字段的仿真数据,并将仿真数据与真实数据进行相似性检验,生成的仿真数据与真实数据符合程度极高;同时,由于政府制定的政策会对社会的个体行为产生很大的影响,从而导致仿真数据与真实数据的偏差过大。因此,在研究仿真数据的同时,也应该关注相应的政策制度,将政策仿真与数据仿真相结合,使得仿真数据根据政策进行自动调整。对社会保险参保缴费行为进行分析,利用经典博弈模型分析缴费行为中各个参与人的博弈关系,并采用博弈演化的思想,提出一个基于进化博弈的参保缴费行为模型,并分析了该模型下不同政策参数对个体缴费行为以及社保机构的影响;最后,以此模型为基础对养老金缴费基数生成能够根据政策的变化进行自我调整的、具有一定自适应性的仿真数据,构建完整的仿真数据库。
[Abstract]:In order to ensure the rapid and stable development of our country's economy, audit institutions should give full play to their functions, which requires them to be good at forecasting, identifying and defusing economic risks, and establishing a forecast of audit risk awareness. This paper finds out the main influence factors and problems on the economic development from the complicated problems, and reports the feedback information to the relevant departments of the country in time, and puts forward the suggestions of policy audit. These are all based on audit data, but the current situation is that the extraction of audit data needs a lot of manpower, material resources, time, and some of the obtained data has a certain degree of confidentiality and heterogeneity. It can not be used well by auditors, which makes it difficult to carry out some audit services. Therefore, it is necessary to build a simulation database. The research content of this paper is mainly aimed at the pension insurance in the field of social insurance. Based on the nine table 153 fields in the national standard library of the endowment insurance industry, the paper analyzes the relationship between the fields of the pension insurance industry and the pension insurance industry tables. By processing and fitting the real data, the simulation data of 50 fields in 6 data tables are generated by Monte Carlo method, and the similarity between the simulation data and the real data is verified. At the same time, the policy made by the government will have a great impact on the individual behavior of the society, which leads to the deviation between the simulation data and the real data. Therefore, when studying the simulation data, we should also pay attention to the corresponding policy system, combine the policy simulation with the data simulation, make the simulation data adjust automatically according to the policy. Based on the analysis of social insurance contributory behavior, this paper uses the classical game model to analyze the game relationship of each participant in the contributory behavior, and puts forward a model based on evolutionary game. It also analyzes the influence of different policy parameters on individual contribution behavior and social security institutions under this model. Finally, based on this model, the generation of pension contribution base can be self-adjusted according to the changes of policy. A complete simulation database is constructed with adaptive simulation data.
【学位授予单位】：哈尔滨工程大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F842.67;F239.4;TP391.9

【参考文献】