ABSTRACT

Social scientists are often concerned with how several variables are related to each other. Economic theory may say something about how the money supply and inflation are related and an econometrician may want to test that theory on the basis of some data. This chapter provides a discussion on linear regression model. The assumption of normality is a restrictive one, but the general idea is that the mean of one variable can be modeled as a linear function of outcomes of other variables. The chapter illustrates the estimation of parameters of a linear regression model by using the least squares method. It discusses the decomposition of errors caused by the ordinary least squares approach.