ABSTRACT
This chapter discusses: a distinguishing feature of a multiple linear regression additive model; the interpretation of the Model F -test for this model; R 2, adjusted R 2, root mean square, and the coefficient of variation as measures of fit of a multiple linear regression model to the data; partial t-tests; the role of collinearity when the results from a Model F -test and partial t- tests are conflicting; why a confidence interval estimate for a regression coefficient of a model predictor provides more information than a partial t-test for the regression coefficient for that predictor; the multiplicity issue and partial t-tests of model predictors; the multiplicity issue associated with confidence and prediction intervals in multiple linear regression; and detection of hidden extrapolation in multiple linear regression.
