A consulting firm is expanding to a new location. To estimate the

amount of office space it needs in this new location, it performed the following analysis of

the quantity of office space used by existing offices in 60 locations. The explanatory

variable Number of Employees is a count of the number of staff working in the office, and

the response Office Space (sq ft) is the amount of office space at the location, in square feet

12. The fitted model estimates that an office staffed and equipped like these with 25

employees would have, with 95% probability assuming the SRM, from

a) 4,415 to 9,655 square feet.

b) 5,725 to 8,345 square feet.

c) 7,004 to 7,102 square feet.

d) 5,870 to 8,200 square feet.

e) 5,705 to 9,365 square feet.

13. If the point in the plot marked with "x" were removed from estimation and the

regression equation refit, then we can be assured that

a) The R2 statistic would become larger.

b) The RMSE would remain 1,310.

c) The standard error of the slope would increase.

d) The estimated intercept b0 would decrease.

e) The estimated slope b1 would increase.

14. Assuming the SRM holds, if an office like these were to expand by 5 employees, then

the fitted model estimates that, with 95% confidence, the office will need

a) From about 821 to 1,311 additional square feet.

b) From about 1,020 to 1,120 additional square feet.

c) From about 1,040 to 1,090 additional square feet.

d) Up to more than 3,680 additional square feet.

e) An amount of additional space that depends on the size of the office

15. The fitted equation suggests that a typical office that is set up like these

a) Uses about 1,700 square feet for storage, bathrooms, and common areas.

b) Has a total of 1,700 square feet.

c) Gives every employee 215 square feet of space for an office.

d) Has a total of about 1,915 employees.

e) Dedicates 215 square feet to storage areas and open areas.

16. The R2 statistic of the fitted model implies that

a) The fitted model accurately predicts the space needed by about 56% of offices.

b) There is about a 56% chance that the fitted model is the true model.

c) The correlation between the predicted values and response is about 0.56.

d) The fitted equation accounts for more than of the variation in office size.

e) About 56% of the observed data lie within 1,310 sq ft of the fitted equation.

17. If the response variable were expressed in square meters rather than square feet (1

square meter = 10.764 square feet), then

a) The R2 statistic would increase by about 0.11.

b) The RMSE would be divided by 10.764.

c) The estimated intercept would remain 1,702.

d) The estimated slope would be multiplied by 10.764.

e) The t-statistic for the slope would be divided by 10.764.

18. In order to confirm that the data in this example are a suitable match to the assumption

of equal error variation in the Simple Regression Model, we should inspect the

a) Leverage plot for Number of Employees.

b) Histogram of the response Office Space.

c) Histogram and boxplot of the residuals.

d) Normal quantile plot of the residuals.

e) Scatterplot of the residuals versus Number of Employees.

19. A review of company records uncovered another 30 offices that are organized like

these. Assume that the same SRM applies to the initial sample of 60 and these

additional 30. In a regression of Office Space on Number of Employees using all 90

cases, we can be assured that

a) The R2 statistic would be larger than 0.5665.

b) The RMSE would be larger than 1,310.

c) The estimated slope would be statistically significantly larger than 213.

d) The standard error of the slope would be smaller than 24.5.

e) The estimated intercept would be equal to 1,702.486.