ABSTRACT
This chapter discusses the assumptions of Classical Linear Regression Model (CLRM) with reference to independent variables and error term to show that the estimation technique, OLS has a number of desirable properties (BLUE), and the hypothesis tests for the coefficient estimates could validly be conducted. These assumptions of CLRM can be used to handle the twin problems of statistical inference, namely, estimation and hypothesis testing, as well as the problem of prediction. The assumptions are ‘linearity’ in regression model, variability in X values, specification of the regression model, there is no perfect multicollinearity, and no autocorrelation between the error terms or disturbances.
