Multicollinearity: A Problem in Multiple Linear Regression
Vaibhav Chittora*
Dr. YSPUHF, Nauni, Solan, Himachal Pradesh (173 230), India
Heerendra Prasad
Dr. YSPUHF, Nauni, Solan, Himachal Pradesh (173 230), India
Prashant Vasishth
ICAR-Indian Agricultural Research Institute, Pusa, New Delhi, Delhi (110 012), India
Mohit Sharma
ICAR-Indian Agricultural Research Institute, Pusa, New Delhi, Delhi (110 012), India
DOI: NIL
Keywords: Correlation, Matrix, MLR, VIF
Abstract
In regression analysis it is obvious to have a relation between the response and regressor(s) variables, but having linear relation among regressor variables is an undesired thing. Multicollinearity refers to the linear relation among two or more variables. If this happens, the standard error of the coefficients will increase. It is a data problem that may cause serious difficulty with the reliability of the estimates of the model parameters. Multicollinearity makes some variables statistically insignificant when they should be significant. In this article, we focus on the multicollinearity, reasons, and consequences of the reliability of the regression model.
Downloads
not found
Reference
Montgomery, D.C., Peck, E.A., Vining, G.G., 2014. Introduction to linear regression analysis. WILEY, Singapore, p. 325.
Greene, W.H., 2003. Econometric Analysis. Prentice Hall, New Jersey, p. 56.
Gujarati, D.N., 2005. Basic Econometrics. McGraw-Hill, New York, p. 341.
Daoud, J.I., 2017. Multicollinearity and Regression Analysis. Journal of Physics: Conf. Ser. 949, 012009.