Question: Use the dosage/pulse rate data in problem 3.59, page 152 of the text for this question.

a. First, fit an intercept model to the data and perform a full analysis by following the regression modeling steps (i.e. scatter plot, testing of assumptions, residual plots, confidence and prediction intervals, as well as ANOVA and t test on coefficients).

b. Then, fit a no-intercept model and perform a full analysis.

c. Has the quality of the model fit decreased using a no intercept model? Which is more useful?