基于主成分分析对新零售销售量灰色预测GM(1,1)模型

GM(1,1) Model for Grey Prediction of New Retail Sales Based on Principal Component Analysis

基于预先处理得到的前九个月销售量数据性质离散∞,通过建立灰色预测GM(1,1)模型,预测后三个月的销售量并进行误差分析。再结合主成分分析,通过降维来更好地突出模型的优点,提出主成分回归–灰色预测模型,减少了回归分析所需要考虑的变量个数,并使预测结果具有较好的准确性,具有较好应用前景。本文结合主成分分析、灰色预测的优点,通过降维将粗糙但相关性高的成分组合,预测销售量,给出进货建议。

Based on the discretization of the first nine months sales data obtained by pre-processing, the grey forecast GM(1,1) model is established to predict the sales volume in the last three months and to carry out error analysis. Combined with principal component analysis, the advantages of the model are better highlighted by dimensionality reduction, and the principal component regression-gray prediction model is proposed to reduce the number of variables to be considered in regression analysis, and to make the prediction results more accurate. It has a good application prospect. This paper combines the advantages of principal component analysis and grey prediction to predict sales volume and give purchase suggestions by reducing dimension to combine rough but highly correlated components.