论文部分内容阅读
财政收支趋势预测具有重要应用价值。本文建立了财政收入回归支持向量机的时间预测模型,并以贵州地方财政收入为例,对采集到的数据进行归一化处理后,分别采用RBF核函数和Linear核函数的支持向量机和时间序列模型对数据进行应用,以1980-2001年的数据为拟合数据,以2002-2007年的数据为预测数据,比较相对误差,得出采用径向量核函数的支持向量机回归预测的结果准确度高。
Forecasting the fiscal revenue and expenditure has important application value. In this paper, a time prediction model of fiscal revenue regression support vector machine is established. Taking the local financial revenue in Guizhou as an example, the collected data are normalized and processed by using RBF kernel function and Linear kernel function support vector machine and time The model was applied to the data. The data from 1980 to 2001 were used as the fitting data and the data from 2002 to 2007 were used as the forecasting data. The relative errors were compared, and the results of the regression prediction of the SVM with radial kernel function were accurate High degree.