摘要
在现有财税政策背景下,以河南省为例,选取15个可能影响财政收入的因子,分别使用逐步回归选择变量-多元线性回归、岭回归以及自适应Lasso建立了模型,分别分析了影响河南省地方财政收入的关键因素,最终将三种方法进行对比。结果显示,使用自适应Lasso建立的模型拟合效果最好。
Under the background of existing fiscal and tax policies,taking Henan Province as an example,15 factors that may affect fiscal revenue were selected,and the model was established by using step-up regression selection variables-multiple linear regression,ridge regression and adaptive Lasso.The key factors affecting local fiscal revenue in Henan Province were analyzed respectively,and the three methods were compared finally.The results show that the model established by using adaptive Lasso has the best fitting effect.
作者
刘怡
徐平峰
LIU Yi;XU Pingfeng(School of Mathematics and Statistics,Changchun University of Technology,Changchun 130012,China)
出处
《长春工业大学学报》
2023年第3期232-238,共7页
Journal of Changchun University of Technology
基金
吉林省自然科学基金项目(20210101152JC)。