欢迎访问林业科学,今天是

林业科学 ›› 2004, Vol. 40 ›› Issue (2): 75-78.doi: 10.11707/j.1001-7488.20040213

• 论文及研究报告 • 上一篇    下一篇

用两阶段度量误差模型方法和ForStat软件进行模型整合

李永慈 唐守正 李海奎   

  1. 北京林业大学资源环境学院,北京100083;中国林业科学研究院资源信息研究所,北京100091
  • 收稿日期:2002-08-26 修回日期:1900-01-01 出版日期:2004-03-25 发布日期:2004-03-25

Model Integration Using the Method of Two-stage Error-in-variable and the Soft of ForStat

Li Yongci,Tang Shouzheng,Li Haikui   

  1. College of Resource and Environment, Beijing Forestry University Beijing100083;The Institute of Forest Information Research, Chinese Academy of Forestry Beijing100091
  • Received:2002-08-26 Revised:1900-01-01 Online:2004-03-25 Published:2004-03-25

摘要:

以同龄林分的年龄X、平均单株胸径Y1(cm)和平均单株材积Y2(m3)的关系模型为例,应用ForStat软件,分别用常规方法和两阶段度量误差模型方法对模型进行整合。结果表明,应用常规方法得到的Y2 估计值明显有偏,而应用度量误差模型方法得到的Y2估计值明显优于应用常规方法得到的估计结果。事实证明:度量误差模型方法和ForStat软件为模型整合提供了有效的途径和工具

关键词: 模型整合, 两阶段度量误差模型, 数值算法

Abstract:

In this study, The models of stand age, stand mean diameter at breast height and stand volume are integrated by using the method of two-stage error-in-variable and the ordinary method separately with the soft of ForStat. The result indicates that the estimation of stand volume obtained by ordinary method has bias while the estimation of two-stage error-in-variable is obviously more accurate and has no bias. It is proved that the method of the two-stage error-in-variable and the soft of ForStat are useful method of integrating models.

Key words: Model integration, Two-stage error-in-variable, Numerical algorithm