论文部分内容阅读
林业生产中所用材积表是建立在一个回归模型的基础上的,由材积表可得到相应于一组自变量(如胸径、树高等)单株立木的平均材积,进一步可估算出一块或几块标准地的蓄积量,以至于估算面积已知的整个林区的总蓄积量。但对于这种估计在一定概率保证下的置信区间却一直研究很少。本文介绍了线性回归分析中对自变量的多次取值求相应因变量之和置信区间的原理和方法。这个方法最早是由 O’REGAN 提出来的。本文利用这种方法对陕西省永寿县槐坪林场人工刺槐林的多株材积之和求出了一个置信区间。与实际测得的资料相比较,预报多株材积之和取得了令人满意的精度。
The volume tables used in forestry production are based on a regression model from which the average volume of stand trees per tree for a set of independent variables (eg, diameter at breast height and height of tree) can be obtained and one or more The amount of stockpiles in the standard so that the total volume of the entire forest area with the estimated area is known. However, there has been little research on the confidence interval of this kind of estimation under a certain probability guarantee. This paper introduces the principle and method of determining the sum of confidence intervals of the dependent variables for multiple values of independent variables in linear regression analysis. This method was first proposed by O’REGAN. In this paper, we use this method to calculate the confidence interval of the sum of multiple plant products of artificial Robinia pseudoacacia forest in Huaiping Forest Farm, Yongshou County, Shaanxi Province. Compared with the actual measured data, the prediction of the sum of multi-product volume achieved satisfactory accuracy.