REGULARIZED LEAST-SQUARE OPTIMIZATION METHOD FOR VARIABLE SELECTION IN REGRESSION MODELS

Authors

  • Marko Dimovski Ss. Cyril and Methodius University in Skopje image/svg+xml Author
  • Irena Stojkovska Ss. Cyril and Methodius University in Skopje image/svg+xml Author

DOI:

https://doi.org/10.37560/matbil11700080d

Keywords:

linear regression, regression models, least square method, regularization, penalty functions

Abstract

A new type of regularization in least-square optimization for variable selection in regression models is proposed. Proposed regularization is suitable for regression models with equal or at least comparable regressors’ influence. Consistency of the estimator of the regression parameter under suitable assumptions is shown. Numerical results demonstrate efficiency of the proposed regularization and its better performance compared to existing regularization methods.

Downloads

Published

2017-01-01

Issue

Vol. 41 No. 1 (2017): 41 (LXVII) 1 (2017)

Section

Articles

License

Copyright (c) 2017 Matematichki Bilten

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

How to Cite

[1]
M. Dimovski and I. Stojkovska, “REGULARIZED LEAST-SQUARE OPTIMIZATION METHOD FOR VARIABLE SELECTION IN REGRESSION MODELS”, Mat. Bilt., vol. 41, no. 1, pp. 80–100, Jan. 2017, doi: 10.37560/matbil11700080d.