Measure Value Future Period Forecast
Error Measures 9. 1,362,143.
Bias (Mean Error) -0.0156 10. 1,455,952.
MAD (Mean Absolute Deviation) 50,773.7969 11. 1,549,762.
MSE (Mean Squared Error) 3,498,808,832. 12. 1,643,572.
Standard Error (denom=n-2=6) 68,301.3828 13. 1,737,381.
Regression line 14. 1,831,191.
Demand (y) = 517857.2 15. 1,925,000.
+ 93,809.5234 * Time (x) 16. 2,018,810.
Statistics 17. 2,112,619.
Correlation coefficient 0.9642 18. 2,206,429.
Coefficient of determination (r^2) 0.9296 19. 2,300,238.
20. 2,394,048.
21. 2,487,857.
Case- kwik Lube
Question# 1 compute the loss for Kwik Lube stations during the last two years using regression. How accurate can the results claim to be?
Question # 2 Was it worth $ 20000 to perform the marketing research?
Question # 3 What other factors might be introduced into the lawsuit?
Dick Johnson, an owner of the Kwik lube company, has hired DR. Gunn to file a lawsuit against T.A Williams after he had violated a Franchise contract with Kwik Lube. To file a lawsuit Dr. Gunn was trying to find data about the same industry or a similar one in a location a location resembling the area in which the original problem occurred. So, Dr. Gunn decided to collect a data fro LA area. This would require the development of the questionnaire that could determine the total gross number of cars serviced for fast oil and lubrication business in the Los Angeles area between 1980 and 1990.
Answer for # 1 question:
Kwik Lube- Regression Model
Demand Y LA(X)
1 680,000 220000
2 750,000 250000
3 750,000 240000
4 780,000 260000
5 990,000 330000
6 1,040,000 350000
7 1,200,000 390000
8 1,330,000 440000
Created by QM for Windows
The following table shows the data (1981 - 1988) that went into calculating trend and regression Models.
According to the above data correlation of coefficient is .9642 and coefficient of determination is .9296 because both models have high percent of accuracy. Sales for