小波变换-DGM(2,1)预测全国税收收入

Forecast of National Tax Revenue Based on the Wavelet DGM (2,1) Model

我国税收收入数据序列呈波浪式上升趋势,传统的预测方法难以获得理想的预测效果。小波变换可以去除原始信号中的噪声,灰色DGM (2,1)模型适合波动时间序列的预测。本文运用小波变换和灰色DGM (2,1)模型相结合的方法预测我国的税收收入,并取得了理想的效果,2017年的预测误差仅为-0.6070%,比灰色DGM (1,1)模型的8.5975%减小92.9398%,比普通灰色DGM (2,1)模型的1.6649%减小63.5414%。本文由此可预测得到2018年我国税收收入为13.025万亿元。

The data sequence of tax revenue in China is rising in a wave style.The traditional forecast method is difficult to obtain an ideal prediction effect.The wavelet transform can remove the noise in the original signal,and the grey DGM(2,1)model is suitable for the prediction of the time series.In this paper,the method of combining the wavelet transform with the grey DGM(2,1)model is used to predict the income of China’s tax revenue,and the ideal results have been achieved.The prediction error in 2017 is only-0.6070%,which is 92.9398%less than the result of 8.5975%by the grey GM(1,1)model,and is 63.5414%less than the result of 1.6649% by the normal grey DGM(2,1)model.It is predicted by the model that the tax revenue of China in 2018 is 13.025 Trillion Yuan.