用麦夸方法最优拟合逻辑斯谛曲线

一般对非线性逻辑斯谛生长曲线的拟合是采用首先对原方程线性化以后用线性最小二乘的方法。此方法不是最优的。本文提出用麦夸方法对曲线进行拟合,并且比较了Gause、Andrewartha、May、Pearl、Krebs、万昌秀等人提出的方法与计算结果。麦夸方法对生物实验及生态学中诸多非线性曲线的参数估计具有普遍的意义。

FITTING LOGISITIC CURVE BY MARQUARDT'S ALGORITHM

To transform the nonlinear equation to provide a mathematically equivalent linear equation is a common method for fitting logisitic equation in the past. But it will not yield the minimum residuals for the untransformed equation. In this paper, Marquardt's algorithm is used for determining the values of parameters in logisitic equation that will give the best least-squares fit to a set of data.The comparison of present method with the methods, used by Gause,And-rewartha, May, and Wan, are given in the paper...