Multiple Imputation for Missing Data in Inventory on Dominant Tree Height

摘要：

In the paper we discuss the application of MI in the missing data analysis. Using MCMC method of Bayesian principle and the proc MI and proc Mianalyze of SAS software we make multiple imputations for missing data of forest resource investigation and get the result analysis in different losing-rate. Eventually we draw the conclusion that MI methods is a valid tool for missing data of forest resource investigation.

关键词： Missing Data Parameter Estimation Multiple Imputation MCMC

作者: ZHANG Xiangyun SONG Hongfeng

作者单位: Zhejiang Forestry College, Linan，Zhejiang Province, 311300

会议类型: 国际会议

会议名称: 2009 International Institute of Applied Statistics Studies(2009 国际应用统计学术研讨会)

会议地点: 青岛

会议语种:英文

页码: 1-6

在线出版日期: 2009-07-25（万方平台首次上网日期，不代表论文的发表时间）