Height–diameter Equations of Three Main Coniferous Tree Species in Changbai Mountain, China Using Mized-effects Modeling Approach
Statistical models in which both fixed and random effects enter nonlinearly are becoming increasingly popular. These models have a wide variety of applications. A new SAS/STATR procedure, NLMIXED, fits these models using likelihood-based methods. Height-diameter relationships were investigated for Spruce (Picea koraiensis Nakai), Fir (Abies holophylla Maxim.) and Korean Pine (Pinus koraiensis Sieb. Et Zucc.) using data from uneven-aged mixed forest in Changbai Mountain area, China. Chapman-Richards functions were fitted to the height-diameter models. A mixed-effect modeling approach was applied in fitting the models for tree species by using PROC NLMIXED. Heights predicted by models including random effects parameters (calibrated response) were compared with those developed without random-effects parameters (fixed-effects response). Results indicate that including the random parameter consistently resulted in a better fit to the data (smaller AIC values).
PROC NLMIXED Growth Model Application Chapman-Richards Functions
ZHAO Junhui KANG Xingang
The Key Laboratory for Silviculture and Conservation of Ministry of Education, Beijing Forestry University, Beijing, P.R.China, 100083
国际会议
2009 International Institute of Applied Statistics Studies(2009 国际应用统计学术研讨会)
青岛
英文
1-5
2009-07-25（万方平台首次上网日期，不代表论文的发表时间）