ABSTRACT

This chapter considers linear regression models (or simply linear models) that are undoubtedly the most important and used class of models. It exemplifies how the various general results on statistical theory can be applied to this more complex and realistic model. Linear models appear in regression analysis when one explores the impact of one or a group of variables called independent variables, predictors, explanatory variables, inputs, factors, covariates, etc. on another (quantitive) variable called a dependent variable, response, output, etc. The common method for estimating the unknown regression coefficients is the method of least squares. Prediction is one of the main goals in regression analysis. Analysis of variance (ANOVA) is undoubtedly one of the most widely used statistical techniques. The simplest ANOVA model is one-way ANOVA that involves a single factor, e.g., the variance in the results of a medical test for various types of drugs.