本文主要研究内容
作者（2019）在《Simulation of the growth ring density of Larix olgensis plantation wood with the ARMA model》一文中研究指出：Because growth ring data have temporal features, time series analysis can be used to simulate and reveal changes in the life of a tree and contribute to plantation management. In this study, the autoregressive(AR) and moving average modeling method was used to simulate the time series for growth ring density in a larch plantation with different initial planting densities. We adopted the Box–Jenkins method for the modeling, which was initially based on an intuitive analysis of sequence graphs followed by the augmented Dickey–Fuller stationarity test. The order p and q of the ARMA(p, q) model was determined based on the autocorrelation and partial correlation coefficient figure truncated on the respective order.Through the residual judgment, the model AR(2) was only fitted to the larch growth ring density series for the plantation with the 1.5 9 2.0 m~2 initial planting density.Because the residuals series for the other three series was not shown as a white noise sequence, the modeling was rerun. Larch wood from the initial planting density of2.0 9 2.0 m~2 was modeled by ARMA(2, 1), and ARMA((1, 5), 3) fitted to the 2.5 9 2.5 m~2 initial planting density,and the 3.0 9 3.0 m~2 was modeled by AR(1, 2, 5).Although the ARMA modeling can simulate the change in growth ring density, data for the different growth ring time series were described by different models. Thus, time series modeling can be suitable for growth ring data analysis, revealing the time domain and frequency domain of growth ring data.
Abstract
Because growth ring data have temporal features, time series analysis can be used to simulate and reveal changes in the life of a tree and contribute to plantation management. In this study, the autoregressive(AR) and moving average modeling method was used to simulate the time series for growth ring density in a larch plantation with different initial planting densities. We adopted the Box–Jenkins method for the modeling, which was initially based on an intuitive analysis of sequence graphs followed by the augmented Dickey–Fuller stationarity test. The order p and q of the ARMA(p, q) model was determined based on the autocorrelation and partial correlation coefficient figure truncated on the respective order.Through the residual judgment, the model AR(2) was only fitted to the larch growth ring density series for the plantation with the 1.5 9 2.0 m~2 initial planting density.Because the residuals series for the other three series was not shown as a white noise sequence, the modeling was rerun. Larch wood from the initial planting density of2.0 9 2.0 m~2 was modeled by ARMA(2, 1), and ARMA((1, 5), 3) fitted to the 2.5 9 2.5 m~2 initial planting density,and the 3.0 9 3.0 m~2 was modeled by AR(1, 2, 5).Although the ARMA modeling can simulate the change in growth ring density, data for the different growth ring time series were described by different models. Thus, time series modeling can be suitable for growth ring data analysis, revealing the time domain and frequency domain of growth ring data.
论文参考文献
论文详细介绍
论文作者分别是来自Journal of Forestry Research的，发表于刊物Journal of Forestry Research2019年02期论文，是一篇关于,Journal of Forestry Research2019年02期论文的文章。本文可供学术参考使用，各位学者可以免费参考阅读下载，文章观点不代表本站观点，资料来自Journal of Forestry Research2019年02期论文网站，若本站收录的文献无意侵犯了您的著作版权，请联系我们删除。