Page 261 - Applied Statistics with R
P. 261

Chapter 13




                      Model Diagnostics







                           “Your assumptions are your windows on the world. Scrub them off
                           every once in a while, or the light won’t come in.”
                           — Isaac Asimov

                      After reading this chapter you will be able to:

                         • Understand the assumptions of a regression model.
                         • Assess regression model assumptions using visualizations and tests.
                         • Understand leverage, outliers, and influential points.
                         • Be able to identify unusual observations in regression models.


                      13.1     Model Assumptions


                      Recall the multiple linear regression model that we have defined.


                               =    +       +       + ⋯ +      −1   (  −1)  +    ,     = 1, 2, … ,   .
                                                             
                                  0
                                      1   1
                                             2   2
                                                                      
                               
                      Using matrix notation, this model can be written much more succinctly as
                                                      =      +   .
                      Given data, we found the estimates for the    parameters using


                                                  ̂
                                                       ⊤
                                                               ⊤
                                                   = (     ) −1       .
                      We then noted that these estimates had mean
                                                       261
   256   257   258   259   260   261   262   263   264   265   266