两水平非线性混合模型对杉木林优势高生长量研究

利用两水平非线性混合模型对杉木(Cunninghamia lanceolata)优势高进行分析。概述了两水平非线性混合模型并简单介绍了该模型的参数估计方法;选用了5种常见的Richards和Logistic 形式模型作为构建混合模型的基础模型,利用建模数据分别对这些基础模型各自衍生出的19种混合模型进行计算及比较,结果表明:这5种基础模型对应的最佳混合模型分别为模型(3-1) 模型(3-5);最后把这些最佳混合模型及传统的回归模型两两进行比较,结果表明:二水平非线性混合模型拟合效果比传统的回归模型拟合效果要好,并且基础模型4对应的二水平混合模型(式3-4)拟合效果最好。

Study of the Dominant Height for Chinese Fir Plantation Using Two-Level Nonlinear Mixed Effects Model

Nonlinear mixed effects model (NLMEM) is built on the relationship of the fixed and random effects in the regression function. The NLMEM has an obvious comparative advantage in analyzing the longitudinal data, repeated measures data and multilevel data. Two-level NLMEM is used to analyze the dominant height for Chinese fir (Cunninghamia lanceolata). The authors outline the two-level NLMEM and introduce the parameters estimation method of the model. Based on five common Richard and Logistic models, the mixed model is built. The modeling data are used to calculate and compare with 19 models derived from each based model, and 5 optimal mixed models are built. Compared the 5 optimal mixed models with traditional regression models, it is showed that the two-level NLMEM has a better fitting effect than the regression models.