摘要
正交最小一乘方法由于其稳健性而在工程中有广泛的应用,然而求解线性模型正交最小一乘参数估计算法往往过于复杂或者只对样本和变量个数较少的问题适用.把正交最小一乘参数估计问题转化为组合优化问题,再使用门限接受算法求解,通过计算机仿真说明了本文算法的正确性和有效性.
The least orthogonal absolute deviation criteria is widely used in engineering because of its robustness, but the algorithms for solving the least orthogonal absolute deviation estimate of the regression coefficient either too explicated or only efficient for small samples and variables. In this paper, a new method based on threshold accepting algorithm to solve the least orthogonal deviation estimates of regression coefficient is proposed by changing the problem to the combinatorial optimization based on it＇s properties. At last the numerical experimentations verified the validity of the new method.
出处
《数学的实践与认识》
CSCD
北大核心
2009年第20期122-128,共7页
Mathematics in Practice and Theory
基金
中国地震局教师科研基金(20090126)
防灾科技学院防灾减灾青年基金(2008A05)
关键词
正交最小一乘
门限接受算法
线性模型
least orthogonal absolute deviation
threshold accepting algorithm
linear model
