Burger King items. Like many fast-food restaurant chains, Burger King (BK) provides data on the nutrition content of its menu items on its website. Here’s a multiple regression predicting calories for Burger King foods from Protein content (g), Total Fat (g), Carbohydrate (g), and Sodium (mg) per serving.

Response variable is: Calories R squared = 99.9%

R squared (adjusted) = 99.9%

s = 8.752 with 122 - 4 = 118 degrees of freedom Source Sum of Squares df Mean Square F-ratio Regression 6434653 3 2144884 28002 Residual 9038.61 118 76.5984 Variable Coefficient SE(Coeff) t-ratio P-value Intercept 1.55770 1.965 0.793 0.4296 Protein(g) 3.86017 0.0952 40.6 60.0001 Fat(g) 9.22763 0.0863 107 60.0001 Carbs(g) 3.87753 0.0467 83.1 60.0001

a) Do you think this model would do a good job of predicting calories for a new BK menu item? Why or why not?

b) The mean of Calories is 452.1 with a standard deviation of 230.8. Discuss what the value of s in the regression means about how well the model fits the data.

c) Does the R2 value of 99.9% mean that the residuals are all actually equal to zero?