Doing least squares: Perspectives from Gauss and Yule

被引:33
作者
Aldrich, J [1 ]
机构
[1] Univ Southampton, Dept Econ, Southampton SO17 1BJ, Hants, England
关键词
least squares; Gauss; Yule; Fisher; Aitken; elimination; partial correlation; orthogonalisation; projection;
D O I
10.2307/1403657
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
Gauss introduced a procedure for calculating least squares estimates and their precisions. Yule introduced a new system of notation adapted to correlation analysis. This paper describes these formalisms and compares them with the matrix and vector space formalisms used in modern regression analysis.
引用
收藏
页码:61 / 81
页数:21
相关论文
共 95 条
[21]  
CHEBYSHEV PL, 1859, MEM ACAD IMPER SCI S, V7, P1
[22]  
CHEBYSHEV PL, OEUVRES, V1
[23]  
COCHRAN W. G., 1938, J. Roy. Statist. Soc., (Suppl.), V5, P171
[24]   The distribution of quadratic forms in a normal system, with applications to the analysis of covariance. [J].
Cochran, WG .
PROCEEDINGS OF THE CAMBRIDGE PHILOSOPHICAL SOCIETY, 1934, 30 :178-191
[25]   Note on the independence of certain quadratic forms [J].
Craig, AT .
ANNALS OF MATHEMATICAL STATISTICS, 1943, 14 :195-197
[26]  
David F.N., 1938, Stat. Res. Mem, V2, P105
[27]  
Davidson R.D., 1993, ESTIMATION INFERENCE, DOI DOI 10.1017/S0266466600009452
[28]  
Deming W. E., 1943, STAT ADJUSTMENT DATA
[29]  
DEMPSTER A.P., 1969, ELEMENTS CONTINUOUS
[30]  
Dieudonne Jean, 1981, History of Functional Analysis.