组合预测模型在税收收入预测中的应用

Application of Combination Forecasting Method in Tax Revenue Forecasting

准确预测税收收入,对于有效地进行市场经济条件下的宏观调控有着重要的意义.为了充分利用各单项预测的信息以提高预测精度,在现有研究的基础上,首先选取指数平滑法、偏最小二乘方法和灰色预测方法对税收收入数据进行定量分析;然后基于误差平方和最小构建了税收收入组合预测模型;最后的算例预测结果表明,构建的税收收入组合预测模型具有较优的预测效果.

Forecasting the tax revenue exactly is significant to carry on the macroscopic regulation efficiently under the market. To take full advantage of the information of each single prediction model to improve forecasting results, the article, on the basis of existing research, carries out quantitative analysis on tax revenue by using exponential smoothing method, partial least squares method and the grey prediction method. Then the combined forecasting model of tax revenue is constructed based on the minimum sum of squared errors. Finally the case study verity that the combined forecasting model put forward in this paper can reduce errors and improve the prediction accuracy.