林分干材收获模型及其弱复共线性的克服

以林分干材收获模型建模的实例，提出与文献有异的共线性诊断标准，并以简取繁，进行旨在克服弱复共线性的岭估计，建立岭回归方程。     

城市森林绿地和环境质量定量评价的研究*

本文以国家统计部门公开发表的资料为基础,选用全国11个大城市的有关16个因子进行评价分析。用标准复回归方法量定各因素对环境的影响,并采用聚类分析和主成分分析方法,以验证定量评定的可信性,其结果与环境评价模型基本一致。     

基于主成分回归及遥感的贵阳市林分郁闭度估测

以贵阳市为研究区,以Landsat-8 OLI为遥感信息源,通过偏相关分析,选择了与郁闭度相关的12个遥感因子作为自变量因子。为克服自变量因子间严重的多重共线性和模型的不稳定性,采用主成分回归分析法建立郁闭度估测模型,并对模型进了检验及精度验证。结果表明,回归方程调整后的R2=0.756,模型的拟合效果较好,说明模型在数学上是可行的;利用15个实测样地对模型进行精度验证,估测精度的平均水平为78.80%,说明该模型可为区域林分郁闭度估测提供参考。     

用岭估计研究以RS和GIS为基础的森林郁闭度估测

本文根据最小二乘估计（LS）及岭估计（R）原理，利用平均残差平方和准则，从样地对应的RS和GIS信息中，用计算机仿真方法筛选出影响郁闭度估测的主要信息。在分析LS估计存在缺陷的基础上，提出采用岭估计建立郁闭度估测方程。实例分析表明，当在影响郁闭度估测的变量间存在复共线性时，岭估计比LS估计具有较高的预报精度，可以满足生产的需要。     

木香花挥发性组分的定量结构-色谱保留关系研究

采用分子电性距离矢量(MEDV)表征了木香花挥发性成分的分子结构,建立了色谱定量结构-保留关系(QSRR)模型,同时运用多元线性回归/逐步回归对模型进行变量筛选,建立了53种木香花挥发性成分气相色谱保留时间(tR)与MEDV的定量相关10 变量、 6变量模型及线性保留指数(ILR)的10、7、6变量模型.QSRR模型的建模计算值复相关系数(R)分别为0.949、0.906、0.906、0.949和0.943.留一法交互校验复相关系数(RCV)分别为0.867、0.904、0.905、0.901和0.904.结果表明所建模型具有良好估计能力与稳定性.     

基于GF-1与Landsat-8的康保县叶面积指数遥感反演研究

以GF-1和Landsat8遥感影像为数据源,采用逐步回归、非线性Logistic回归和基于空间位置的地理加权回归3种方法,结合134个野外样地调查数据,在河北省康保县开展叶面积指数反演研究,并对结果进行精度检验。结果表明:(1)在荒漠化地区,GF-1和Landsat-8遥感影像提取的植被指数因子与LAI均有较高的相关性。运用主成分分析方法对植被指数因子进行处理,可以有效消除各影响因子间的共线性。(2)基于GF-1和Landsat-8影像分别建立的3种模型,均以地理加权回归决定系数最大,均方根误差最小,反演精度最高。(3)国产GF-1数据反演LAI效果优于Landsat-8,可以代替Landsat-8数据进行叶面积指数的估测。     

通径分析在林业多元线性模型共线性分析中的应用研究

以思茅松多元线性树高模型为例,应用通径分析方法对该模型中的共线性进行了分析,并对比分析了相关系数分析方法和条件指数、容限度与方差膨胀因子判定方法.研究结果表明,通径分析方法不仅能判断林业多元线性模型中是否存在多重共线性,而且还能判断其多重共线性的不同来源和强弱;相对于常用的多重共线性检验方法,通径分析方法更具优越性.     

上层冠层因子对马尾松肉桂复层林下层林分生物量的影响

采用标准地定位监测方法,选取不同栽培模式马尾松(Pinus massoniana)肉桂(Cinnamomum cassia)人工复层林上层林冠的叶面积指数、平均叶倾角、直接辐射透过系数、散射辐射透过系数、消光系数等5个冠层因子作为影响下层林分生长的光环境指标,利用主成分分析法研究影响肉桂林分生物量的主要光环境影响因子。结果表明,不同上层密度光环境下的肉桂各器官生物量变化明显;在主成分分析中,马尾松冠层叶面积指数、平均叶倾角和直接辐射透过系数三者的方差贡献率累计达到99.37%,是影响下层林生长的主导光环境因子;肉桂生物量的主成分回归模型具有良好的拟合度,从回归模型中可得出影响肉桂生物量的主要因子是上层林冠层的直接辐射透过系数。     

森林资源调查监测年度数据产出方法研究

利用2021年国家林草生态综合监测评价北京市样地数据及相关资料,基于“1/5+4/5”联合估计方案,提出回归抽样估计和回归模型预测2种基本估计方法;同时还针对部分省份存在多个副总体和加密样地等特殊情况,提出了采用分层抽样和联合估计方法产出年度数据,并以北京市森林蓄积量估计为例,以4套方案开展计算。结果显示,基于老样地估计的2套方案,全市森林蓄积量估计精度均在92%以上,其中方案2(基于回归模型预测)的总体抽样精度从方案1(基于回归抽样估计)的92.87%提高到93.87%;采用加密新样地和老样地联合估计的2套方案,全市森林蓄积量估计精度均在94%以上,其中方案4的总体抽样精度从方案3的94.24%提高到94.65%。研究表明：采用回归抽样估计和回归模型预测2种基本方法对当年未调查样地进行数据更新的方法,在实践中是完全可行的;综合利用回归抽样估计、回归模型预测、分层抽样方法和联合估计方法,可以解决存在副总体和加密样地等不同情况下的年度数据产出问题。     

最小二乘递推估计与一次估计的一致性

回归模型是随机模型的框架性模型，随机差分方程及Ｋａｌｍａ滤波等模型均是回归模型框架中的特殊模型。弄清回归模型的最小二乘递推估计与一次估计的一致性，也就弄清了随机差分方程及Ｋａｌｍａ滤波等模型的递推估计的实质。为此，本文论述最小二乘递推估计与一次估计的一致性。     

Relationship between height growth of Scots pine (Pinus sylvestris L.) and site factors in Great Britain

The rate of height growth in Scots pine stands throughout Great Britain was examined in relation to site factors. The site factors included measures of geographical variation, topography, soil chemical and physical variables at two levels, several measures of soil phosphorus status, foliar monoterpenes and estimates of mean values of climatic variables. Principal component analysis was used to make an initial selection of regressor variables from 88 site factors for use in multiple regression equations. A dependent variable was obtained by removing the effect of age on height. The dependence of this expression on site variables was then examined using step-up multiple regression, starting with the selection of regressor variables made by the principal component analysis (P.C.A.), then adding transformations and interaction terms until 69% of variation was significantly accounted for over Great Britain, and up to 99% for parts of the country. Simpler equations using only those regressor variables which could be read from maps were also calculated.Variations in growth over Great Britain are associated mainly with solar radiation, soil texture and soil moisture content. For separate parts of the country the solar radiation term disappears from regression equations.     

影响南平市木材价格主要因子的分析研究

本研究遵循资料易于获取的原则，从统计年鉴中选取可能影响木材价格的因素，采用主成分分析法和逐步回归分析法进行分析。研究结果表明：影响木材价格的主要因素，即亩蓄积、林地利用率、人均林业总产值和当地物价指数。     

Additive biomass equations for small diameter trees of temperate mixed deciduous forests

The carbon stored in small diameter trees of temperate forests has been ignored in most studies and there is a lack of biomass equations for this component of forests. We harvested nine main tree species at sapling stage (dbh?相似文献   

体育木地板龙骨结构与主要性能的回归统计关系

以高等级竞赛用体育木地板为研究对象,采用测定体育木地板主要功能指标的方法,研究不同龙骨参数的体育木地板结构性能,建立体育木地板龙骨结构(截面尺寸、间距)与冲击吸收率(Fr)、标准垂直变形(Vd)和球反弹率(Br)之间的一次回归方程,分析提出较优的龙骨设计参数。结果表明:龙骨参数对试验指标均有显著影响,显著性水平均为0.05及以上,综合分析各功能指标与试验因素水平组合,得出较优的龙骨设计参数为截面尺寸50mm×50mm、龙骨间距500mm×500mm。该研究填补体育木地板龙骨结构与主要性能指标间一次回归统计关系的空白,建立的统计方程可在理论统计范畴内指导体育木地板中龙骨结构设计及施工技术,为体育木地板龙骨的生产提供科学依据。     

Measurement of lumber moisture content based on PCA and GS-SVM

Lumber moisture content(LMC) is the important parameter to judge the dryness of lumber and the quality of wooden products.Nevertheless the data acquired are mostly redundant and incomplete because of the complexity of the course of drying,by interference factors that exist in the dryness environment and by the physical characteristics of the lumber itself.To improve the measuring accuracy and reliability of LMC,the optimal support vector machine(SVM) algorithm was put forward for regression analysis LMC.Environmental factors such as air temperature and relative humidity were considered,the data of which were extracted with the principle component analysis method.The regression and prediction of SVM was optimized based on the grid search(GS) technique.Groups of data were sampled and analyzed,and simulation comparison of forecasting performance shows that the main component data were extracted to speed up the convergence rate of the optimum algorithm.The GS-SVM shows a better performance in solving the LMC measuring and forecasting problem.     

SYSTEMATIC STUDY OF PHYSICAL AND MECHANICAL PROPERTIES OF IMPORTANT CHINESE WOODS

In this paper,the physical and mechanical properties of important Chinese woodswere studied by using the methods of variable analysis, regression analysis,correlation analysis,principle component analysis and so on systematically.The results showed synthetical characteristicof the physical and mechanical properties of Chinese woods.Some opinions of former authors wererevised and some new laws were discovered.That is helpful to study of the physical and mechanicalproperties of woods.     

基于主成分、聚类与SVR组合算法的森林生物量估算方法研究

以孟家岗林场二类清查数据为基础,对1371个小班的11项指标进行主成分分析,并采用系统聚类法对小班进行分类,进而利用支持向量回归算法分别进行生物量模型训练.结果表明：7个主成分指标可反映87.995％的生物量信息;1371个小班可分为5类,各类训练模型的预测精度均在89％以上,且均以v-SVR模型为最优.在得到的5类生物量训练模型基础上估算林场森林乔木层生物量,无需分起源、树种、立地类型,能够在保证生物量估算精度的同时,大大减少工作量,可为区域生物量的估算提供一种新的方法.     

不同立地条件对毛竹生长影响研究

以咸宁市咸安地区40块毛竹标准地的调查资料为基础,分析了坡向、坡度、坡位、腐质层厚度、郁闭度和立株度对毛竹胸径生长量的影响。首先对这6个立地因子进行逐步回归分析,结果表明坡度、腐质层厚度和坡位是影响毛竹胸径生长的主要因子。再对6个立地因子进行主成分分析,其2个主成分累积贡献率达78.5%,并计算出各标准地的这2个主分量值。最后在主成分分析结果的基础上,对40块标准地进行了聚类。     

Applicability of Fitted and Published Biomass Equations for Larix Species in Northeastern China

ABSTRACT

This study reviewed published biomass equations for Larix species in the literature and developed additive biomass equations (by nonlinear seemingly unrelated regression known as NSUR) and independent equations, respectively, based on DBH-only and DBH-H variables with the biomass data of 150 trees sampled in northeastern China. The applicability of fitted and published equations was comprehensively assessed with respect to Larix species, their provinces and origins when applied in northeastern China. The results showed that equations derived from the local study area did not always perform better than the published biomass equations in the literature. Tree-level equations performed slightly broader than stand-level equations. Nonlinear regression could provide higher applicability due to correction factors in the linear regression. Adding tree height as the second variable improved the applicability of stem and aboveground published equations. The applicability of original published equations was better for northeastern China for sites of similar latitude. When using published biomass equations, care should be taken to assess the uncertainty and applicability of biomass equations, which is important for the sustainable management of forested resources. Published information on biomass equations is especially necessary for forestry organizations that do not have the opportunity to develop their own equations.     

呼和浩特市经济增长与环境污染的关系分析

从经济增长与环境污染之间关系研究的意义出发,以人均GDP和环境污染指标为研究指标,运用相关分析、回归分析、主成分分析、主成分投影法等方法,对呼和浩特市1991-2004年间的经济增长与环境污染的关系进行定量分析,并建立了呼和浩特市的环境库兹涅茨曲线模型。结果显示：呼和浩特市的环境EKC曲线不符合典型的EKC特征,而呈现出二次曲线的特点,表明呼和浩特市在经济增长过程中尚未经过污染排放量最大点,存在持续上升的态势,呈现出＂边增长、边污染＂的发展模式。因此,为了经济长远发展,必须控制污染。