Multicollinearity
The term multicollinearity was coined in 1934 by Ranger Frisch in his book ‘Confluence
Analysis ‘. Strictly speaking, Multicollinearity refers existence of more than one exact linear
relationships and collinearity refers to existence of single linear relationship. But this
distinction is rarely maintained in practice and multicollinearity refers to both cases. That
means the existence of 'perfect' or exact linear relationship among some or all
explanatory variables of regression model is termed as multicollinearity. It is commonly
associated with multiple regression analysis.
Usually economic variables are related in several ways and Because of the strong relationship
among explanatory variable, the problem of multicollinearity is said to be exist and it become
difficult to find out how much each of these will influence the dependent variable.
Consider the following regression model,
Y = B0+ B1X1 + B2X2 + B3X3+ Ui if the explanatory or independent variable X1, X2,
X3are interrelated, then the problem of multicollinearity is said to exist.
The presence of multicollinearity is shown in the following venn diagram
In the below figure, the independent variables x1 and X2 are not related with each other. So
there is no chance for the presence of multicollinearity.
Y
X1 X2
The presence of multicollinearity is shown in the following venn diagram
Y
X1 X2
Here the independent variable or explanatory variable not only depends on dependent or
explained variable but also they are interrelated with each other.
, X1 X2
The term multicollinearity was coined in 1934 by Ranger Frisch in his book ‘Confluence
Analysis ‘. Strictly speaking, Multicollinearity refers existence of more than one exact linear
relationships and collinearity refers to existence of single linear relationship. But this
distinction is rarely maintained in practice and multicollinearity refers to both cases. That
means the existence of 'perfect' or exact linear relationship among some or all
explanatory variables of regression model is termed as multicollinearity. It is commonly
associated with multiple regression analysis.
Usually economic variables are related in several ways and Because of the strong relationship
among explanatory variable, the problem of multicollinearity is said to be exist and it become
difficult to find out how much each of these will influence the dependent variable.
Consider the following regression model,
Y = B0+ B1X1 + B2X2 + B3X3+ Ui if the explanatory or independent variable X1, X2,
X3are interrelated, then the problem of multicollinearity is said to exist.
The presence of multicollinearity is shown in the following venn diagram
In the below figure, the independent variables x1 and X2 are not related with each other. So
there is no chance for the presence of multicollinearity.
Y
X1 X2
The presence of multicollinearity is shown in the following venn diagram
Y
X1 X2
Here the independent variable or explanatory variable not only depends on dependent or
explained variable but also they are interrelated with each other.
, X1 X2
