×

Multiple criteria linear regression. (English) Zbl 1131.62308

Summary: The unknown parameters in multiple linear regression models may be estimated using any one of a number of criteria such as the minimization of the sum of squared errors MSSE, the minimization of the sum of absolute errors MSAE, and the minimization of the maximum absolute error MMAE. At present, the MSSE or the least squares criterion continues to be the most popular. However, at times the choice of a criterion is not clear from statistical, practical or other considerations. Under such circumstances, it may be more appropriate to use multiple criteria rather than a single criterion to estimate the unknown parameters in a multiple linear regression model. We motivate the use of multiple criteria estimation in linear regression models with an example, propose a few models, and outline a solution procedure.

MSC:

62J05 Linear regression; mixed models
62F10 Point estimation
90C05 Linear programming
90C90 Applications of mathematical programming

Software:

AS 135
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Andre, C. D.S.; Elian, S. N.; Narula, S. C.; Tavares, R. A., Coefficient of determination for variable selection in the MSAE regression, Communications in Statistics—Theory and Methods, 29, 623-642 (2000) · Zbl 1061.62540
[2] Andre, C. D.S.; Narula, S. C.; Elian, S. N.; Tavares, R. A., An overview of the variable selection methods for the minimum sum of absolute errors regression, Statistics in Medicine, 22, 2101-2111 (2003)
[3] Armstrong, R. D.; Kung, D. S., Minimax estimates for a linear multiple regression problem, Applied Statistics, 28, 93-100 (1979) · Zbl 0454.65090
[4] Armstrong, R. D.; Frome, E. L.; Kung, D. S., A revised simplex algorithm for the absolute deviation curve fitting problem, Communications in Statistics: Simulation and Computation B, 8, 2, 175-190 (1979) · Zbl 0439.65049
[5] Arthanari, T. S.; Dodge, Y., Mathematical Programming in Statistics (1981), John Wiley and Sons, Inc.: John Wiley and Sons, Inc. New York, NY · Zbl 0549.62002
[6] Barrodale, I.; Phillips, C., An improved algorithm for discrete Chebychev linear approximations, (Proceedings of 4th Manitoba Conference on Numerical Mathematics (1974), University of Manitoba: University of Manitoba Winnipeg, Man.), 177-190
[7] Barrodale, I.; Phillips, C., Solution of an over-determined system of linear equations in the Chebyshev norm, ACM Transactions on Mathematical Software, 1, 264-270 (1975) · Zbl 0309.65015
[8] Barrodale, I.; Roberts, F. D.K., Solution of an over-determined system of equations in the \(L_1\) norm, Communications of the ACM, 17, 319-320 (1974)
[9] Beckman, R. J.; Cook, R. D., Outliers, Technometrics, 25, 119-163 (1983)
[10] Birkes, D.; Dodge, Y., Alternative Methods of Regression (1993), John Wiley and Sons: John Wiley and Sons New York, NY · Zbl 0850.62528
[11] Caples, S. C.; Hanna, M. E.; Premeaux, S. R., Least squares versus absolute value in real estate appraisal, The Appraisal Journal, January, 17-24 (1997)
[12] Coleman, J. W.; Larsen, J. E., Alternative estimation techniques for linear appraisal models, The Appraisal Journal, October, 526-532 (1991)
[13] Coleman, T. F.; Li, Y., A globally and quadratically convergent affine scaling method for linear \(l_1\) problems, Mathematical Programming, 56, 189-222 (1992) · Zbl 0760.90069
[14] Dielman, T. E., Least absolute value regression: Recent contributions, Journal of Statistical Computation and Simulation, 75, 263-286 (2005) · Zbl 1059.62070
[15] Elian, S. N.; Andre, C. D.S.; Narula, S. C., Influence measure for the \(l_1\) regression, Communications in Statistics—Theory and Methods, 29, 837-849 (2000) · Zbl 0993.62059
[16] Hogg, R. V., Adaptive robust procedures: A partial review and some suggestions for future applications and theory, Journal of American Statistical Association, 69, 909-923 (1974) · Zbl 0305.62030
[17] Ihlanfeldt, K. R.; Martinez-Vazquez, J., Alternative value estimates of owner-occupied housing: Evidence on sample selection bias and systematic errors, Journal of Urban Economics, 20, 356-369 (1986)
[18] Isakson, H. R., Using multiple regression in real estate appraisal, The Appraisal Journal, October, 424-430 (2001)
[19] Madsen, K.; Nielsen, H. B., A finite smoothing algorithm for linear \(l_1\) estimation, SIAM Journal on Optimization, 3, 223-235 (1993) · Zbl 0781.65115
[20] Mantel, N., Restricted least squares regression and convex quadratic programming, Technometrics, 11, 763-773 (1969) · Zbl 0183.48002
[21] Narula, S. C.; Saldiva, P. H.N.; Andre, C. D.S.; Elian, S. N.; Ferreira, A. F.; Capelozzi, V., The minimum sum of absolute errors regression: A robust alternative to the least squares regression, Statistics in Medicine, 18, 1401-1417 (1999)
[22] Portnoy, S.; Koenker, R., The Gaussian hare and the Laplacian tortoise: Computability of squared-error versus absolute error estimators, Statistical Science, 12, 279-300 (1997) · Zbl 0955.62608
[23] Ruzinsky, S.; Olsen, E., \(L_1\) and \(l_∞\) minimization via a variant of Karmarkar’s algorithm, IEEE Transactions on Acoustics, Speech, and Signal Processing, 37, 245-253 (1989)
[24] Zhang, Y., Primal-dual interior point approach for computing \(l_1\) solutions and \(l_∞\) solutions of over-determined systems, Journal of Optimization Theory and Applications, 77, 323-341 (1993) · Zbl 0796.49029
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.