Informatics and Applications
2021, Volume 15, Issue 2, pp 6065
METHOD OF STRAIGHTENING DISTORTED DUE TO MULTICOLLINEARITY COEFFICIENTS IN REGRESSION MODELS
Abstract
When constructing regression models, due to the strong multicollinearity of the explanatory variables,
its coefficients are distorted, in particular, their signs, which negatively affects the interpretational qualities of such
regression. This article is devoted to the development of a method of straightening coefficients distorted due to
multicollinearity This method is based on the property of the fully connected linear regression models proposed by
the author. A nonlinear system, which is used to estimate fully connected regressions, is investigated. It is shown
that the solution of this system can be obtained numerically using the method of simple iterations. A method for
choosing unknown lambdaparameters in fully connected regression is proposed. It was found that in multivariate
fully connected models with a strong correlation of all factors, the signs of the coefficients for the variables in the
secondary equation coincide with the corresponding signs of the correlation coefficients. To straighten the distorted
coefficients on the basis of this research, the "Selection B" algorithm was developed. The developed method
of straightening has been successfully demonstrated by the example of modeling Russia's gross domestic product
(GDP).
[+] References (10)
 Gunst, R. F, and J. T. Webster. 1975. Regression analysis
and problems of multicollinearity. Commun. Stat. Theory
4:277292.
 Tamura, R., K. Kobayashi, Y. Takano, R. Miyashiro,
K. Nakata, and T. Matsui. 2017. Best subset selection
for eliminating multicollinearity J. Oper. Res. Soc. Jpn.
60(3):321336.
 Forster, E., and B. Ronz. 1983. Methoden der korrelation und regressionsanalyse. Berlin: Verlag Die Wirtschaft,
1979. 369 p.
 Chatterjee, S., and A. S. Hadi. 2012. Regression analysis
by example. 5th ed. Hoboken, NJ: Wiley. 424 p.
 Jolliffe, I. T. 2002. Principal component analysis. NewYork,
NY: SpringerVerlag. 488 p.
 Hoerl, A. E., and R. W. Kennard. 1970. Ridge regression:
Biased estimation for nonorthogonal problems. Technometrics 12:5567.
 Mokshina, S. I., G. V. Shurshikova, and S. S. Shchekunskikh. 2017. Metod postroeniya soderzhatel'no interpretiruemykh regressionnykh modeley v usloviyakh
mul'tikollinearnosti [The construction method of meaningful interpreted regression models in conditions of multicollinearity]. Sovremennaya ekonomika:problemy i resheniya [Modern Economics: Problems and Solutions] 89(5):8194.
 Bazilevskiy, M. P. 2019. Sintez modeli parnoy lineynoy regressii i prosteyshey EIVmodeli [Synthesis of linear regression model and EIVmodel]. Modelirovanie, optimizatsiya i informatsionnye tekhnologii [Modeling, Optimization and Information Technology] 7(1):170182.
 Bazilevskiy, M. P. 2019. Issledovanie dvukhfaktornoy modeli polnosvyaznoy lineynoy regressii [Investigation of a twofactor fully connected linear regression model]. Modelirovanie, optimizatsiya i informatsionnye tekhnologii [Modeling, Optimization and Information Technology] 7(2):8096.
 Bazilevskiy, M. P. 2020. Mnogofaktornye modeli polnosvyaznoy lineynoy regressii bez ogranicheniy na sootnosheniya dispersiy oshibok peremennykh [Multifactor fully connected linear regression models without constraints to the ratios of variables errors variances]. Informatika i ee Primeneniya  Inform. Appl. 14(2):9297.
[+] About this article
Title
METHOD OF STRAIGHTENING DISTORTED DUE TO MULTICOLLINEARITY COEFFICIENTS IN REGRESSION MODELS
Journal
Informatics and Applications
2021, Volume 15, Issue 2, pp 6065
Cover Date
20210630
DOI
10.14357/19922264210209
Print ISSN
19922264
Publisher
Institute of Informatics Problems, Russian Academy of Sciences
Additional Links
Key words
regression analysis; fully connected linear regression model; multicollinearity; interpretation; numerical
method; GDP of Russia
Authors
M. P. Bazilevskiy
Author Affiliations
Department of Mathematics, Irkutsk State Transport University, 15 Chernyshevskogo Str., Irkutsk 664074, Russian
Federation
