一元整体线性回归预测法的MATLAB程序设计  

MATLAB Programming for Unary Total Linear Regression Model

作　　者：彭友[1] 

机构地区：[1]四川建筑职业技术学院,四川德阳618000

出　　处：《科技广场》2015年第7期39-42,共4页Science Mosaic

摘　　要：本文根据一元整体回归模型的误差方程推导整体最小二乘法计算回归参数和拟合优度的一组公式,并给出整体线性回归预测法迭代计算的MATLAB程序,最后用简单算例验证该方法和程序。结果表明,当自变量也含有随机误差时,本文给出的程序正确,整体估计比最小二乘估计更优、更合理。This paper derives a set of formulas for solving unary linear regression problem and computing goodness of fit with the errors function of unary total linear regression model, and gives a series of code in MATLAB language for solving unary total linear regression problem. A case study is given to validate the program of unary total linear regression model at last. The computed results show that when the independent variables are perturbed by random noise, the given code is correct, and the results from the program are better than LS estimation of unary linear regression model.

关 键 词：一元线性回归 整体最小二乘法 MATLAB程序 施工利润 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]