一种新的小样本建模方法──非线性映象降维建模方法

A NEW METHOD OF CONSTRUCTING A SMALL-SAMPLES MODEL──A DIMENSION REDUCING METHOD OF CONSTRUCTING MODEL BY NONLINEAR IMAGES

本文应用多元统计分析中的距离概念，降低变量空间的维数，以解决建立预测模型时遇到的小样本多变量问题。

In this paper the concept of distance in multivariate statistical analysis is used to reduce the dimensions of variable space so as to reso1ve the contradiction of small samples and multivariates encountered in constructing a prediction model.And a concrete example is given to illustrate the application of this method.