基于灰色模型和支持向量回归的财政收入预测

Fiscal Revenue Prediction Based on Grey Model and Support Vector Regression

随着科学技术的不断发展,大数据应用的越来越普及,已成为提高财政收入的有力工具。本文以1994~2019年数据为依托,借助R统计软件,首先对财政收入、第一产业增加值、工业增加值、建筑业增加值、年末总人口、社会消费品零售总额和受灾面积这六个方面的原始数据进行相关性分析,运用Lasso回归方法识别影响财政收入的关键特征,然后将灰色模型和支持向量回归预测模型相结合,对未来两年的财政收入进行预测,最后对建立的财政收入预测模型进行评价。

With the continuous development of science and technology, the application of big data has become more and more popular, and it has become a powerful tool to increase fiscal revenue. Firstly, the relativity of data from 1994 to 2019 among fiscal revenue, primary industry added value, industrial added value, construction industry added value, total population at the end of the year, total retail sales of consumer goods, and disaster-affected area is analyzed by R software in this article. And using the Lasso regression method to choose the key features that affect fiscal revenue. Then we combine the gray model and the support vector regression prediction model to predict the fiscal revenue for the next two years. Finally, the established fiscal revenue forecast model is evaluated.