选择合适的费用驱动因子是装备参数费用建模的一个关键问题。本文基于灰色系统的“新信息优先原理”提出新信息优先的主成分回归模型,证明了新信息优先的主成分分析模型充分反映新信息的误差。通过实例建立新信息优先的主成分回归模型,并与BP神经网络方法、支持向量机方法的结果进行对比,结果表明本文的方法具有更高的预测精度。 Choosing the right cost driver is a key issue in modeling the cost of equipment parameters. In this paper, we propose a principal component regression model based on the gray system “new information priority principle ”, which proves that the new principal component analysis model that gives priority to the new information fully reflects the error of the new information. Through the example, the principal component regression model with new information priority is established and compared with the BP neural network method and the support vector machine method. The results show that the proposed method has higher prediction accuracy.