The file P16_53.xlsx contains data on pork sales. Price is in dollars per hundred pounds, quantity sold is in billions of pounds, per capita income is in dollars, U.S. population is in millions, and...

The file P16_53.xlsx contains data on pork sales. Price is in dollars per hundred pounds, quantity sold is in billions of pounds, per capita income is in dollars, U.S. population is in millions, and GNP is in billions of dollars.


a. Use the data on all potential explanatory variables to develop a regression equation that could be used to predict the quantity of pork sold during future periods.


b. Find the correlations between all pairs of variables. (We recommend StatTools.) What evidence do you see of multicollinearity? How could this cause a problem with the regression? Run another regression without potentially redundant variables. How does this new regression compare to the one in part a?


c. Suppose that during each of the next two quarters, price is 45, U.S. population is 240, GNP is 2620, and per capita income is 10,000. (These are in the units described previously.) Predict the quantity of pork sold during each of the next two quarters using the equations from parts a and b. How do they compare?