The data set in Table 11.25 show U.S. average retail gasoline prices as of the first week of each month from January 2007 through January 2009. It also shows the world average price of crude oil. The information is from the Energy Information Administration, http://www.eia.doe.gov. Prices for gasoline are in cents per gallon, and for crude oil are in dollars per barrel.
(a) Fit a simple linear regression of y = gasoline price versus x = crude oil price, and calculate the residuals.
(b) Plot each residual versus the residual for the preceding month. Does it seem reasonable that the errors are independent?
(c) Use a statistical software package to compute the Durbin-Watson statistic and its p value, and interpret the result.
(d) If gas prices this month are lower than what is predicted from the regression model, what can we say about gas prices next month?
Already registered? Login
Not Account? Sign up
Enter your email address to reset your password
Back to Login? Click here