作者：刘庆, 邵志新

作者单位：中国测试技术研究院, 四川成都 610021

关键词：回归分析; 直线拟合; 最小二乘法; 参数; 不确定度

摘要：

针对实验数据统计处理问题,一般采用一元线性拟合的基本原理进行回归分析,并由回归直线关系式求出因变量和自变量的定量关系,进而对各个参数的不确定度评估进行探讨。运用实例说明了直线拟合的应用及各参数不确定度评估计算过程,以期在应用过程中加深理解、方便应用。

Discussion of linear fitting uncertainty of regress analysis

LIU Qing, SHAO Zhi-xin

National Institute of Measurement and Testing Technology, Chengdu 610021, China

Abstract: In experimental data statistics analysis, regression often has been done to obtain the quantitative relationships between dependent variables and independent variables based on the principle of single variable linear fitting, orderly to get the estimates of the parameters and their uncertainties. For deeply understanding and easily applying, an example is given to interpret the application of linear fitting and to describe the estimation of the parameters and the evaluation of their uncertainties.

Keywords: Regression analysis; Linear fitting; Method of least squares; Parameters; Uncertainty

2009, 35(3): 41-44  收稿日期: 2008-9-5;收到修改稿日期: 2008-11-17

基金项目: 

作者简介: 刘庆(1970-),女,工程师,主要从事气体分析仪检定校准质检工作。

参考文献

[1] JJF 1059-1999,测量不确定度评定与表示[S].北京:中国计量出版社,1999.
[2] 肖明耀.误差理论与应用[M].北京:中国计量出版社,1985.
[3] 梁晋文.误差理论与数据处理[M].北京:中国计量出版社,1989.
[4] 李慎安.测量不确定度表达百问[M].北京:中国计量出版社,2001.
[5] 张彦立.硫酸亚铁剂量计吸收剂量基准改造后的剂量学性能研究[J].现代测量与实验室管理,2008,16(1):3-8.