THE ACCURACY OF LEAST-SQUARES CALCULATIONS WITH THE CHOLESKY ALGORITHM

被引:5
作者
RANDALL, JH [1 ]
RAYNER, AA [1 ]
机构
[1] UNIV NATAL,DEPT STAT & BIOMETRY,PIETERMARITZBURG 3200,SOUTH AFRICA
关键词
D O I
10.1016/0024-3795(90)90357-I
中图分类号
O29 [应用数学];
学科分类号
070104 ;
摘要
In a detailed study of the accuracy of the Cholesky method for solving least squares equations, points in the algorithm at which accuracy is lost are identified, methods for improving accuracy are discussed, and simple formulas for predicting accuracy from the maximum variance inflation factor (as well as from the condition number) are evaluated. The central role of cancellation in determining accuracy is emphasized. © 1990.
引用
收藏
页码:463 / 502
页数:40
相关论文
共 61 条
[31]  
KENNEDY WJ, 1980, STATISTICAL COMPUTIN
[32]   LEAST-SQUARES PROGRAMS, A LOOK AT SQUARE ROOT PROCEDURE [J].
KOPITZKE, R ;
BOARDMAN, TJ ;
GRAYBILL, FA .
AMERICAN STATISTICIAN, 1975, 29 (01) :64-66
[33]  
LESAGE JP, 1988, J STATIST COMPUT SIM, V30, P273
[34]   LEAST-SQUARES COMPUTATIONS AND THE CONDITION OF THE MATRIX [J].
LONGLEY, JW .
COMMUNICATIONS IN STATISTICS PART B-SIMULATION AND COMPUTATION, 1981, 10 (06) :593-615
[36]  
LONGLEY JW, 1984, LEAST SQUARES COMPUT
[37]  
Maindonald J. H., 1977, Journal of Statistical Computation and Simulation, V5, P247, DOI 10.1080/00949657708810155
[38]  
MAINDONALD JH, 1976, AM STAT, V30, P202
[39]  
MARQUARDT DW, 1980, J AM STAT ASSOC, V75, P87, DOI 10.2307/2287388
[40]   GENERALIZED INVERSES, RIDGE REGRESSION, BIASED LINEAR ESTIMATION, AND NONLINEAR ESTIMATION [J].
MARQUARDT, DW .
TECHNOMETRICS, 1970, 12 (03) :591-+