基于多元线性回归模型的税收影响因素实证研究
An Empirical Study of Tax Influencing Factors Based on Multiple Linear Regression Models

税收是政府财政收入的基本要素，事关国计民生，很有必要洞悉税收的影响因素。本文选取国内生产总值、国民生产总值、财政支出、全社会固定资产投资总额、货币供应量、进口总额和社会消费品零售总额作为指标要素，构建税收收入的多元线性回归模型，并利用Stata软件对我国1990~2019年的实际数据做实证研究，分析税收增长的因素。利用多重性、异方差、和自相关对模型进行检验和修正，得到影响税收因素的显著因素，为政府税收部门决策提供参考。
Taxation is a fundamental element of government fiscal revenue, which is related to national economy and people’s livelihood. It is necessary to understand the influencing factors of taxation. This paper selects gross domestic product, gross national product, fiscal expenditure, total fixed assets investment of the whole society, money supply, total imports and total retail sales of social consumer goods as indicator elements, builds a multiple linear regression model of tax revenue, and uses Stata software to do empirical research on China’s actual data from 1990 to 2019 to analyze the factors of tax growth. Using multiplicity, heteroscedasticity, and autocorre-lation to test and correct the model, significant factors affecting tax factors are obtained, providing reference for government tax department decision-making.