Data 3.4 on page 209 describes a sample of n = 25 Mustang cars being offered for sale on the Internet. We would like to predict the Price of used Mustangs (in $1000s) and the possible explanatory variables in MustangPrice are the Age in years and Miles driven (in 1000s).
(a) Fit a simple linear model for Price based on Age. Does Age appear to be an effective predictor of Price? Justify your answer.
(b) Fit a multiple regression model for Price based on Age and Miles. Is Age an effective predictor in this model? Justify your answer.
(c) Can you think of an explanation for the change from (a) to (b)?

Data 3.4 on page 209

A statistics student, Gabe McBride, was interested in prices for used Mustang cars being offered for sale on an Internet site. He sampled 25 cars from the website and recorded the age (in years), mileage (in thousands of miles), and asking price (in $1000s) for each car in his sample. The data are stored in MustangPrice and the scatterplot in Figure 3.26 shows the relationship between the Miles on each car and the Price. Not surprisingly, we see a strong negative association showing the price of a used Mustang tends to be lower if it has been driven for more miles. The correlation between Price and Miles for this sample is r = ˆ’0.825.

Figure 3.26

Transcribed Image Text:

50 - 40 - 30 - 20 - 10 20 40 80 140 60 100 120 160 Miles Price