logistic模型中饱和值确定方法的研究  

Research on the Ascertainment Method of Parameter Saturated Value in Logistic Model

作　　者：李爱民[1] 刘杉[2] 吕安民[3] 

机构地区：[1]郑州大学水利与环境学院,河南郑州450001 [2]河南省国土资源科学研究院,河南郑州450008 [3]郑州市国土资源局,河南郑州450006

出　　处：《郑州大学学报（理学版）》2011年第1期118-121,共4页Journal of Zhengzhou University:Natural Science Edition

基　　金：国土资源部2006"百名优秀青年科技人才计划"项目

摘　　要：logistic模型是一种重要的非线性预测模型,但如何确定饱和值L一直未得到很好解决.在学习以往方法的基础上,提出了一种新方法,即利用SPSS软件对logistic模型进行L值的选优计算,以模型计算值与样本值的标准离差是否趋于稳定作为值选取的标准,并以郑州市建成区非农人口的预测为例,说明了该方法在人口预测logistic模型中的具体应用.Logistic model was the important non-linear model of time series forecasting.But how to ascertain the parameter saturated value of logistic model had been not well solved till now.A kind of new method was put forward on the base of the former methods.Some Logistic models were primarily made up with different parameter by SPSS,then whether the standard deviation between the forecast value with the real value became steady or not was the criteria of ascertaining the parameter finally.Prediction of nonagricultural population of built-up area in Zhengzhou city was taken as an example.It explained application value of this method.

关 键 词：LOGISTIC模型 饱和值 非农人口 预测 

分 类 号：O242.1[理学—计算数学]