Multicollinearity represents the very high intercorrelations or inter-association among the independent (or predictor) variables.
Multicollinearity takes place when independent variables of multiple regression analysis are highly associated with each other. This association is caused by a high correlation among independent variables. This high correlation will trigger a problem in the linear regression model prediction results. It's the basic assumption of linear regression analysis to avoid multicollinearity for better results:
- It occurs due to the inappropriate use of dummy variables.
- It also occurs due to the repetition of similar variables.
- It is also caused due to synthesized variables from other variables in the data.
- It can occur due to high correlation among variables.
Multicollinearity causes the following problems:
- It causes difficulty in estimating the regression coefficients precisely and coefficients become more susceptible to minor...
