武器型号费用参数模型的数据诊断与识别

针对采用回归分析建立武器型号费用参数模型时，如何选择合适的数据诊断与识别方法进行了讨论。重点介绍了原始数据中多重共线性的判别和如何识别自变量、因变量异常值的方法。采取这些措施后，可确保模型所用数据满足要求，从而提高模型的精度，为进一步建立合理的武器型号费用参数模型，提供可靠保障。

DATA DIAGNOSIS AND DISCRIMINATION FOR THE WEAPON TYPES' COST PARAMETER MODEL

This paper discussed the methods which were used for diagnosing and distinguishing data when the regression analysis was adopted to establish the weapon types' cost parameter model. The methods of distinguishing the collinearity of firsthand data and the value out of the ordinary from independent variable and dependent variable, were especially presented. Thus the data were guaranteed to fit the demand of the model, and the precision of this model was improved.