Abstract
The solution of the matricial systemZ=HX+V is derived. The observations are given by the matrixZ(n×m) and arem-components vectors. The parameter matrix isH(n×g), whileX(g×m) is the unknown andV(n×m) the residual matrix. The solutionX is estimated in order to minimize anL p -norm ofV defined with the help of anL r -norm on theV rows. The minimization algorithm is a steepest descent method with a specially chosen step size to “avoid” the singularities. The general problem shrinks in simple regression or location problems whilem org is reduced to the unit.
Possibly more important is the recursive solution of this same general problem; a solutionX (n) being obtained, we require the solutionX (n+1) of the same system plus one row. In this context the productHX can be seen as a filtered version of the originalZ. The inversion of a system ofg linear equations is required at each step of the recursive solution.
Similar content being viewed by others
References
N. N. Abdelmalek,Linear L 1 approximation for a discrete point set and L 1 solutions of overdetermined linear equations, J. Ass. Comp. Mach 18 (1971), 41–47.
D. F. Andrews, P. J. Bickel, F. R. Hampel, P. J. Huber, W. H. Rogers, and J. W. Tuckey,Robust estimates of location, Princeton Univ. Pr. (1972).
H. Ekblom,Calculation of linear best L p approximations, BIT 13 (1973), 292–300.
A. B. Forsythe,Robust estimation of straight line regression coefficients by minimizing p-th power deviations, Technometrics 14 (1972), 159–166.
W. M. Gentleman,Robust estimation of multivariate location by minimizing p-th power deviations, Thesis at Princeton University, and Memorandum MM 65-1215-16, Bell Tel. Labs. (1965).
F. R. Hampel,The influence curve and its role in robust estimation, J. A. S. A., 69 (1974), 383–393.
P. J. Huber,Robust statistics: a review, Ann. Math. Stat. 43 (1972), 1041–1067.
A. H. Jazwinski,Stochastic processes and filtering theory, Academic Press (1970).
G. Merle and H. Späth,Computational experiences with discrete L p approximation, Computing 12 (1974), 315–321.
W. Rey,On least p-th power in multiple regressions and location estimations, Report R 261, MBLE Res. Lab., Brussels (1974).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Rey, W. On leastp-th power methods in multiple regressions and location estimations. BIT 15, 174–184 (1975). https://doi.org/10.1007/BF01932691
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF01932691