Abstract
The problems of finding the best one-dimensional and multidimensional linear models are considered for a short selection of measurement results with a rather small number of erroneous measurements. Using an example, it is shown that the least moduli method allows one to find models more adequate to the process than the least squares method.
Similar content being viewed by others
REFERENCES
V. N. Vapnik (ed.), Algorithms and Programs of Dependence Restoring [in Russian], Nauka, Moscow (1984).
P. Hewber, Robustness in Statistics [Russian translation], Mir, Moscow (1984).
V. I. Zorkal'tsev, The Least Squares Method [in Russian], Nauka, Novosibirsk (1995).
P. I. Stetsyuk, Yu. G. Kolesnik, and M. M. Leibovich, “On robustness of the method of least squares,” Komp. Mat., No. 1, 1–9 (2002).
O. V. Babak, “An approach to problem solution in a class of piecewise linear functions,” Probl. Upravl. Inform., No. 6, 134–141 (1995).
O. V. Babak, “A principle of self-organizing mathematical models,” Probl. Upravl. Inform., No. 2, 98–107 (2001).
N.Z. Shor, P. I. Stetsyuk, and M. M. Leibovich, “Selecting a method of estimating linear model parameters,” in: Proc. 9th Intern. Acad. M. Kravchuk Conf. (May 16-19, 2002, Kiev) [in Ukrainian], NTTU KPI, Kiev (2002), p. 399.
N. Z. Shor, Minimization Methods for Nondifferentiable Functions and their Applications [in Russian], Naukova Dumka, Kiev (1979).
A. G. Ivakhnenko and I. A. Müller, Self-Organizing Predictive Models [in Russian], Kiev Tekhnika (1985).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Babak, O.V., Gasanov, A.S., Leibovich, M.M. et al. An Approach to Application of the Least Moduli Method in Constructing Linear Models. Cybernetics and Systems Analysis 40, 400–406 (2004). https://doi.org/10.1023/B:CASA.0000041998.15178.84
Issue Date:
DOI: https://doi.org/10.1023/B:CASA.0000041998.15178.84