再荐误差的β分布统示法  被引量:7

Recommendation again for the \%β\% distribution uniform expression method of error

在线阅读下载全文

作  者:林洪桦[1] 

机构地区:[1]北京理工大学,北京100081

出  处:《中国计量学院学报》2004年第2期96-101,共6页Journal of China Jiliang University

摘  要: 论述误差β分布统示法的必要性、可行性和实用性,并再次予以推荐.众所周知,误差分布并不限于正态分布,许多实际情况下需用非正态分布,甚至是非对称分布来描述.β分布统示法恰能胜任这样的描述.提出β分布参数的"界似"和"形似"的两种自助(bootstrap)估计,分布界限估计,异常值剔除及其稳健性剔除等方法均表明β分布统示法具有可行性和实用性.In this paper, the \%β\% distribution uniform expression method of error was recommended again, and its necessity, feasibility and practicability were discussed. Know as well, the error distribution is not limited in normal distribution, and actually, it usually is nonnormal distribution in many cases, even described by assymmetric distribution, The \%β\% distribution uniform expression method is just appropriate to the description of this error distribution. Some methods have been pointed out in the paper, such as two kinds of 'boundary similarity' and 'shape similarity' bootstrap estimates about the parameter of the \%β\% distribution, distributed boundary estimation, elimination to the outlier and its robustness method. Each of them can prove that the \%β\% distribution uniform expression method of error is feasible and practicable.

关 键 词:误差分析 β分布统示法 自助估计法 分布界限估计 异常值稳健性剔除 非正态分布 

分 类 号:O212[理学—概率论与数理统计]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象