论文部分内容阅读
高校科研项目暂付款管理水平关系着资金使用效率,甚至影响科研项目会计信息的真实性。本文根据某省属高校科研项目2006-2015年暂付款及影响暂付款规模的科研收入、科研支出、差旅费、专用材料费、设备购置费5个主要因素数据,建立BP人工神经网络预测模型,预测暂付款规模,并与传统多元线性回归模型比较。数据研究表明BP神经网络能捕捉到暂付款与影响因素间非线性规律,能更好预测科研项目暂付款规模。
The scientific research project temporary payment management of colleges and universities related to the efficiency of the use of funds, and even affect the authenticity of scientific research project accounting information. According to the data of five major factors such as temporary payment of scientific research project of universities and colleges in 2006-2015 and the amount of temporary payment affecting scientific research, scientific research expense, travel expenses, special materials cost and equipment purchase cost, a BP artificial neural network prediction model is established. Temporary payment scale, and compared with the traditional multiple linear regression model. Data studies show that BP neural network can capture the non-linear law between the provisional payment and the influencing factors and can better forecast the temporary payment of scientific research projects.