Standard Linear Regression

This chapter discusses standard linear regression model. In the standard linear regression model, the response y is a continuous measurement variable such as sales or profit. Virtually all statistical packages include easy-to-use routines for the estimation of regression models. Their output provides estimates of the regression coefficients, standard errors of the estimated coefficients, summary statistics about the model fit, and predicted values and prediction intervals for new cases. Strategies for simplifying regression models are also described in texts on regression. Automatic stepwise regression techniques (backward elimination, forward selection, and true stepwise regression) are related methods that simplify models automatically. The R function is used to fit linear (regression) models. The chapter presents R program for measuring fuel efficiency of automobiles, and predicting toyota used-car prices.