Page 42 - FINAL CFA II SLIDES JUNE 2019 DAY 2
P. 42

LOS 8.h: Distinguish between and interpret the                 READING 8: MULTIPLE REGRESSION AND ISSUES IN REGRESSION ANALYSIS
      2
                         2
     R and adjusted R in multiple regression.
                                                                    MODULE 8.4: COEFFICIENT OF DETERMINATION & ADJUSTED R-SQUARED
     COEFFICIENT OF DETERMINATION, R                    2

     In addition to an F-test, the multiple coefficient of determination, R , can be used to test the overall effectiveness of the entire set
                                                                           2
     of independent variables in explaining the dependent variable. Its interpretation is similar to that for simple linear regression: the
     % of variation in the dependent variable that is collectively explained by all of the independent variables.

                         2
     For example, an R of 0.63 indicates that the model, as a whole, explains 63% of the variation in the dependent variable.

                                                        NOTE: Regression output often includes multiple R, which is the correlation
                                                        between actual values of y and forecasted values of y.

                                                        Multiple R is the square root of R . For a regression with one independent variable,
                                                                                           2
                                                        the correlation between the independent variable and dependent variable is the
                                                        same as multiple R (with the same sign as the sign of the slope coefficient).

      Adjusted R         2

      Unfortunately, R by itself may not be a reliable measure of the explanatory power of the multiple regression model. This is
                       2
      because R almost always increases as variables are added to the model, even if the marginal contribution of the new
                  2
      variables is not statistically significant.

                                         2
      Consequently, a relatively high R may reflect the impact of a large set of independent variables rather than how well the set
      explains the dependent variable. This problem is called overestimating the regression.

                                                                                                                             2
       To overcome this the impact of additional variables on the explanatory power of a regression model, we adjust R for the
       number of independent variables.
   37   38   39   40   41   42   43   44