A NOTE ON RECENT PROPOSALS FOR COMPUTING L1 ESTIMATES

被引：7

作者：

BASSETT, GW ^{[1
]}

KOENKER, RW ^{[1
]}

机构：

[1] UNIV ILLINOIS,DEPT ECON,CHICAGO,IL 60680

来源：

COMPUTATIONAL STATISTICS & DATA ANALYSIS | 1992年 / 14卷 / 02期

关键词：

LEAST ABSOLUTE VALUE; LINEAR PROGRAMMING; LEAST SQUARES;

D O I：

10.1016/0167-9473(92)90174-E

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

Recent methods for computing the least absolute value (l1) estimate have been proposed. In contrast to the usual linear programming formulation of the l1 problem, the new methods attempt to use least squares residuals to identify observations whose l1 residuals are equal to zero. Examples are presented to show that the proposals do not produce estimates that are identical or even necessarily close to l1, and hence the algorithms cannot be recommended as a method for computing the l1 estimate.

引用

收藏

页码：207 / 211

页数：5

相关论文

共 11 条

[1] SOLUTION OF AN OVERDETERMINED SYSTEM OF EQUATIONS IN L1 NORM [F4] [J].

BARRODALE, I ;

ROBERTS, FDK .

COMMUNICATIONS OF THE ACM, 1974, 17 (06) :319-320

[2] LINEARLY CONSTRAINED DISCRETE L1 PROBLEMS [J].

BARTELS, RH ;

CONN, AR .

ACM TRANSACTIONS ON MATHEMATICAL SOFTWARE, 1980, 6 (04) :594-608

[3] ASYMPTOTIC THEORY OF LEAST ABSOLUTE ERROR REGRESSION [J].

BASSETT, G ;

KOENKER, R .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1978, 73 (363) :618-622

[4] DISCUSSION OF AN EXAMPLE SHOWING THAT A NEW TECHNIQUE FOR LAV ESTIMATION BREAKS DOWN IN CERTAIN CASES [J].

CHRISTENSEN, GS ;

SOLIMAN, SA ;

ROUHI, A .

COMPUTATIONAL STATISTICS & DATA ANALYSIS, 1990, 9 (02) :203-213

[5]

GALLANT AR, 1974, COMMENTS COMPUTING M

[6] AN EXAMPLE SHOWING THAT A NEW TECHNIQUE FOR LAV ESTIMATION BREAKS DOWN IN CERTAIN CASES [J].

HERCE, MA .

COMPUTATIONAL STATISTICS & DATA ANALYSIS, 1990, 9 (02) :197-202

[7] A NEW POLYNOMIAL-TIME ALGORITHM FOR LINEAR-PROGRAMMING [J].

KARMARKAR, N .

COMBINATORICA, 1984, 4 (04) :373-395

[8]

MEKETON MS, 1985, UNPUB LEAST ABSOLUTE

[9] ITERATIVE TECHNIQUE FOR ABSOLUTE DEVIATIONS CURVE FITTING [J].

SCHLOSSMACHER, EJ .

JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION, 1973, 68 (344) :857-859

[10] A NEW ALGORITHM FOR NONLINEAR L1-NORM MINIMIZATION WITH NONLINEAR EQUALITY CONSTRAINTS [J].

SOLIMAN, SA ;

CHRISTENSEN, GS ;

ROUHI, AH .

COMPUTATIONAL STATISTICS & DATA ANALYSIS, 1991, 11 (01) :97-109