• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>2008年国际应用统计学术研讨会（2008 International Institute of Applied Statistics Studies）论文集 >Nonlinear Mixed Models for Tree Height Growth
【24h】

Nonlinear Mixed Models for Tree Height Growth

机译：树高增长的非线性混合模型

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

A height growth model was developed for Mongolian pine （Pinus sylvestris L.var. mongolica Litv.) in northeastern China based on simple Logistic growth model using nonlinear mixedeffects modeling approach. The methods of model development involve which parameters should be considered to be random and which should be purely fixed and determination of a autoregressive correlation structure. Model performance was evaluated utilizing information criterion statistics including Likelihood ratio tests (LRT), Akaike information criterion (AIC), and Bayesian information criterion (BIC). The Logistic model with three random parameters showed the best performance. The first-order autoregressive AR (1) model was incorporated into the mixed-effects model to provide significant difference on model performance. Heights predicted by model including random-effects parameters (calibrated prediction) were compared with that developed without random-effects parameters (fixed-effects prediction). Including the random parameters resulted in more accurate height prediction.
机译：建立了蒙古松（Pinus sylvestris L.var。）的身高生长模型。基于非线性混合效应建模方法的简单Logistic增长模型。模型开发方法涉及哪些参数应该被认为是随机的，哪些参数应该是完全固定的以及自回归相关结构的确定。使用包括似然比检验（LRT），Akaike信息准则（AIC）和贝叶斯信息准则（BIC）在内的信息准则统计信息来评估模型性能。具有三个随机参数的Logistic模型显示出最佳性能。一阶自回归AR（1）模型被合并到混合效应模型中，以在模型性能上提供显着差异。将包括随机效应参数的模型预测的高度（校准预测）与没有随机效应参数的模型预测的高度（固定效应预测）进行比较。包括随机参数可导致更精确的高度预测。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号