Informatics and Applications
2022, Volume 16, Issue 4, pp 20-25
GENERALIZATION OF A METHOD FOR STRAIGHTENING COEFFICIENTS DISTORTED DUE TO MULTICOLLINEARITY IN REGRESSION MODELS WITH DIFFERENT DEGREES OF EXPLANATORY VARIABLES CORRELATION
Abstract
When constructing regression models, one of the main problems is multicollinearity. This negative
phenomenon leads to distortion of the regression coefficients, in particular, their signs. Earlier, to solve the
problem of multicollinearity, a method for straightening distorted coefficients was developed which is based on
the construction of a fully connected linear regression model. One of the conditions for its applicability is a close
correlation of absolutely all pairs of explanatory variables. But when solving real applied problems, this condition
is rarely met. Most often, explanatory variables correlate with each other in different ways. The authors propose
a new iterative algorithm for the method of straightening distorted coefficients. A feature of the algorithm is that it
combines the advantages of both traditional multiple models and new fully connected regressions. The developed
algorithm is universal and can be used to construct a regression equation with any structure of the correlation
matrix. The new algorithm has been successfully applied to simulate freight transportation by rail in the Irkutsk
region.
[+] References (10)
- Montgomery, D. C., E. A. Peck, and G. G. Vining. 2012.
Introduction to linear regression analysis. Hoboken, NJ:
John Wiley & Sons. 672 p.
- Jolliffe, I. T, and J. Cadima. 2016. Principal component
analysis: A review and recent developments. Philos. T. Roy. Soc. A 374(2065):20150202. 16 p.
- Naik, G. R. 2018. Advances in principal component analysis: Research and development. Springer. 252 p.
- Kibria, B. M. G., and S. Banik. 2016. Some ridge regression estimators and their performances. J. Modern Applied Statistical Methods 15(1):206-238.
- Hoerl, R. W 2020. Ridge regression: A historical context.
Technometrics 62:420-425.
- Giacalone, M., D. Panarello, and R. Mattera. 2018. Multicollinearity in regression: An efficiency comparison between Lp-norm and least squares estimators. Qual. Quant.
52(4):1831-1859.
- Dawoud, I., and B.M.G. Kibria. 2020. A new biased estimator to combat the multicollinearity of the Gaussian linear regression model. Stats 3(4):526-541.
- Bazilevskiy, M. P. 2020. Mnogofaktornye modeli polnosvyaznoy lineynoy regressii bez ogranicheniy na sootnosheniya dispersiy oshibok peremennykh [Multifactor fully connected linear regression models without con-straints to the ratios of variables errors variances]. Informatika i ee Primeneniya - Inform. Appl. 14(2):92-97.
- Bazilevskiy, M. P. 2021. Metod vypryamleniya iskazhennykh iz-za mul'tikollinearnosti koeffitsientov v regressionnykh modelyakh [Method of straightening distorted due to multicollinearity coefficients in regression models]. Informatika i ee Primeneniya - Inform. Appl. 15(2):60-65.
- Orlova, I. V. 2018. Podkhod k resheniyu problemy mul'tikollinearnosti pri analize vliyaniya faktorov na rezul'tiruyushchuyu peremennuyu v modelyakh regressii [Approach to the solution of the multicollinearity problem at the analysis of the influence of the factors on the resulting variable in models of regression]. Fundamental'nyye issledovaniya [Fundamental Research] 3:58-63.
[+] About this article
Title
GENERALIZATION OF A METHOD FOR STRAIGHTENING COEFFICIENTS DISTORTED DUE TO MULTICOLLINEARITY IN REGRESSION MODELS WITH DIFFERENT DEGREES OF EXPLANATORY VARIABLES CORRELATION
Journal
Informatics and Applications
2022, Volume 16, Issue 4, pp 20-25
Cover Date
2022-12-30
DOI
10.14357/19922264220404
Print ISSN
1992-2264
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
regression analysis; correlation; multicollinearity; method for straightening distorted coefficients; fully connected linear regression model
Authors
M. P. Bazilevskiy
Author Affiliations
Irkutsk State Transport University, 15 Chernyshevskogo Str., Irkutsk 664074, Russian Federation
|