ABSTRACT
This chapter discusses consequences and solutions of multicollinearity and heteroscedastic vs. homoscedastic disturbances. It describes use and interpretation of squared variables; and use and interpretation of the product of two quantitative variables. The multicollinearity arises when explanatory variables are highly correlated. A main objective of employing regression analysis is to show that certain independent variables are significant factors for the dependent variable. However, large standard errors caused by multicollinearity would reduce the significance of possible factors. Concerning the identification of multicollinearity it is important to understand multicollinearity in a continuous way. Multicollinearity is the extent to which an independent variable is correlated with other independent variables. In fact, all observed variables are correlated with each other to some extent. The chapter also discusses econometric issues by stating that: A number of econometric issues in the regression analysis need to be addressed.
