Email this article Solution of the linear regression problem using matrix correction methods in the l1 metric
- Authors: Gorelik V.A.1, Trembacheva (Barkalova) O.S.2
-
Affiliations:
- Dorodnicyn Computing Center
- Moscow State Pedagogical University
- Issue: Vol 56, No 2 (2016)
- Pages: 200-205
- Section: Article
- URL: https://journal-vniispk.ru/0965-5425/article/view/178256
- DOI: https://doi.org/10.1134/S0965542516020081
- ID: 178256
Cite item
Open Access
Access granted
Subscription Access
Abstract
The linear regression problem is considered as an improper interpolation problem. The metric l1 is used to correct (approximate) all the initial data. A probabilistic justification of this metric in the case of the exponential noise distribution is given. The original improper interpolation problem is reduced to a set of a finite number of linear programming problems. The corresponding computational algorithms are implemented in MATLAB.
About the authors
V. A. Gorelik
Dorodnicyn Computing Center
Author for correspondence.
Email: vgor16@mail.ru
Russian Federation, ul. Vavilova 40, Moscow, 119333
O. S. Trembacheva (Barkalova)
Moscow State Pedagogical University
Email: vgor16@mail.ru
Russian Federation, ul. Malaya Pirogovskaya 1, Moscow, 119882
Supplementary files
