基于ARIMAX模型的中国税收收入预测与分析
Prediction and analysis of tax revenue based on ARIMAX Model
赵盼盼 1张迷 2焦力宾3
作者信息
- 1. 营口理工学院 基础教研部,辽宁 营口 115014
- 2. 兰考县统计局,河南 兰考 475399
- 3. 大连理工大学 电子信息与电气工程学部,辽宁 大连 116024
- 折叠
摘要
根据我国 1978-2019 年间的税收收入、国内生产总值和财政支出数据,利用 R软件,对税收收入建立 ARIMA 模型,又以国内生产总值和财政支出为协变量,对税收收入建立 ARIMAX模型.根据两个模型分别预测 2020-2022年税收收入,将预测值与真实值对比,计算得到相对误差和平均绝对百分比误差,发现与 ARIMA 模型相比,ARIMAX 模型误差更小,精度更高,预测更准确.因此,利用ARIMAX模型得到2023-2025年税收收入的预测值.
Abstract
Based on the tax revenue,gross domestic product,and fiscal expenditure data of China from 1978 to 2019,an ARIMA model was established for tax revenue using R software.The ARIMAX model was also established for tax revenue with gross domestic product and fiscal expenditure as covariates.The two models were used to predict tax revenue from 2020 to 2022,and the predicted values were compared with the true values to calculate the relative error and average absolute percentage error.It was found that compared with the ARIMA model,the ARIMAX model has smaller errors,higher accuracy,and more accurate predictions.Therefore,the ARIMAX model is used to obtain the predicted values of tax revenue from 2023 to 2025.
关键词
ARIMAX模型/税收收入/国内生产总值/财政支出/预测Key words
ARIMAX model/tax revenue/gross domestic product/finance expenditure/forecast引用本文复制引用
基金项目
营口理工学院校级科研项目(QNL202025)
出版年
2024
国家科技期刊平台
NSTL
万方数据