Nonlinear Mized Models for Tree Height Growth

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.

关键词： Height Growth Mized-effects Modeling Logistic Model Mongolian Pine

作者: JIANG Lichun

作者单位: Northeast Forestry University, Harbin, P.R.China, 150040

会议类型: 国际会议

会议名称: 2008年国际应用统计学术研讨会(2008 International Institute of Applied Statistics Studies)

会议地点: 烟台

会议语种:英文

页码: 1-5

在线出版日期: 2008-08-14（万方平台首次上网日期，不代表论文的发表时间）