DUE DATE: At the beginning of class on Tuesday December 12,
2000.
NO LATE PROJECTS WILL BE ACCEPTED.
On pages 737 - 740 of the textbook, data are presented for the four
major sports in North America. The goal of the study is to predict
the value of a sports franchise based on various independent
variables.
Listed are the dependent variable (Value of the Franchise), and 8
independent
variables (Franchise Type, Gate Receipts, Stadium Revenues, Player Costs,
Operating Income, Media Revenues, Total Revenues, and Operating
Expenses).
If you are working at Temple, then you can access the data by selecting
‘Open Worksheet’ from the ‘File’ menu in MINITAB. The name of the
data set is SportFrn.dat. You can also enter the data into MINITAB
by hand, or give me a disk and I will copy the data set onto your
disk.
The report that you will hand in should be typed. All MINITAB output
that you want to hand in with the report will be added to the end of the
report. This project is worth 10% of your final grade. The
following is the requirement for the project:
1) Determine the best fitting model to the data. Make the
necessary scatterplots of the data. Go through the procedure of
fitting
the model and removing non-significant variables. Address the issue
of multicollinearity. Make sure that you test the significance of
the regression for your final model. Be sure to include a residual
analysis in your report.
2) Once you have determined the best fitting model, interpret
all of the coefficients in the model. That is, which variables add
to the value of the franchise, and which variables detract from the value
of the franchise. Which variables add more than others, etc.
3) Give the estimated standard deviation of the model and the
coefficient of determination. Interpret the value of the coefficient
of determination in the context of the problem.
4) Predict the value of the franchise using the explanatory
variables
in your model. You can choose the values of the explanatory
variables
for your prediction. Also obtain a 95% prediction interval for this
predicted value and interpret this interval in the context of the problem.