MAXIMUM SUBMATRIX TRACES FOR POSITIVE DEFINITE MATRICES

被引：14

作者：

OLKIN, I ^{[1
]}

RACHEV, ST ^{[1
]}

机构：

[1] UNIV CALIF SANTA BARBARA,DEPT STAT & APPL PROBABIL,SANTA BARBARA,CA 93106

来源：

SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS | 1993年 / 14卷 / 02期

关键词：

MULTIVARIATE DISTRIBUTIONS; COVARIANCE MATRICES; DISTANCE OF RANDOM VECTORS;

D O I：

10.1137/0614027

中图分类号：

O29 [应用数学];

学科分类号：

070104 ;

摘要：

For three p-dimensional jointly distributed random vectors x, y, and z with respective normal marginal distributions N(0, SIGMA(ii)), i = 1, 2, 3, certain covariance matrices are determined that minimize the sum of the L2-distances of the three vectors. This problem posed in a statistical context is equivalent to maximizing submatrix traces of a positive definite matrix.

引用

页码：390 / 397

页数：8

共 31 条

[1] A SCHUR ANALYSIS APPROACH TO MINIMUM DISTANCE PROBLEMS [J].

APITZSCH, W ;

FRITZSCHE, B ;

KIRSTEIN, B .

LINEAR ALGEBRA AND ITS APPLICATIONS, 1990, 141 :105-122

[2] SOME ASYMPTOTIC THEORY FOR THE BOOTSTRAP [J].

BICKEL, PJ ;

FREEDMAN, DA .

ANNALS OF STATISTICS, 1981, 9 (06) :1196-1217

[3]

DARISTOTILE A, 1988, SANKHYA SER A, V50, P363

[4] INVARIANCE-PRINCIPLES IN PROBABILITY FOR TRIANGULAR ARRAYS OF B-VALUED RANDOM VECTORS AND SOME APPLICATIONS [J].

DEACOSTA, A .

ANNALS OF PROBABILITY, 1982, 10 (02) :346-373

[5]

DIACONIS P, 1984, 39 U CAL DEP STAT TE

[6] PRESCRIBING A SYSTEM OF RANDOM VARIABLES BY CONDITIONAL DISTRIBUTIONS [J].

DOBRUSHIN, RL .

THEORY OF PROBILITY AND ITS APPLICATIONS,USSR, 1970, 15 (03) :458-+

[7] THE FRECHET DISTANCE BETWEEN MULTIVARIATE NORMAL-DISTRIBUTIONS [J].

DOWSON, DC ;

LANDAU, BV .

JOURNAL OF MULTIVARIATE ANALYSIS, 1982, 12 (03) :450-455

[8]

DUDLEY RM, 1989, REAL ANAL PROBABILIT

[9]

FRECHET M, 1957, CR HEBD ACAD SCI, V244, P689

[10]

GELBRICH M, 1988, 179 HUMB U BERL SEKT

← 1 2 3 4 →