摘要
本文针对不满足经典回归条件的一对变量，讨论并介绍了非参数Ｍｏｎｏｔｏｎｉｃ回归的建模思想、条件与方法。通过实例说明，Ｍｏｎｏｔｏｎｉｃ回归比最小二乘回归拟合误差小，模型精度高，计算简单，有一定的实用价值。
The article deals with a pair of varables which are not statisfied with traditional regression condition.It also introduces ideas、conditions and methods in building of nonparametric monotonic regression models.Examples have indicated that ditting error of the monotonic regression is less than that of least squares regression.Its models are with precision and caculation,simple.Thus,it is of practial value.
出处
《中国卫生统计》
CSCD
北大核心
1997年第3期1-4,共4页
Chinese Journal of Health Statistics
