几类非线性回归模型的LS估计与非线性度量.pdfVIP

第27 卷第 2 期 湖南文理学院学报(自然科学版) Vol . 27 No. 2 2015 年 6 月 Journal of Hunan University of Arts and Science(Science and Technology) Jun. 2015 doi: 10.3969/j.issn.1672–6146.2015.02.009 几类非线性回归模型的LS估计与非线性度量 肖兵 (湖南文理学院 数学与计算科学学院, 湖南 常德, 415000) 摘要: 研究了一类推广的非线性回归模型的最小二乘估计, 证明了其收敛性, 并推广到不满足高斯– 马尔可夫 条件的情形。研讨了几类特殊的部分非线性回归模型的参数估计及非线性程度度量, 探讨了参数最小二乘估 计求法中迭代初值的选取, 提出了改进方向。 关键词: 非线性度量; 可分离; 局部曲率; LS 估计 中图分类号: O 212.1 文章编号: 1672–6146(2015)02–0026–04 The LSE and measure of non-linearity of some special nonlinear model Xiao Bing (Department of Mathematics and Computational Science, Hunan University of Arts and Science, Changde 415000, China) Abstract: First, the least squares estimator with extend case is studied, and its convergence is proven. Second, the mode to non Gauss-Markov condition is extended. The estimate property of a more generalized separated nonlinear model is discussed. At last, the method of iterative initial value selection are studied, and a new trend for the method to measure the model’s non-linearity is given. Key words: measure of non-linearity; separated; local