Share
Explore BrainMass

# Linear Regression Model

Problem:
We are using a linear regression model (time series) to predict sales of our "WW" brand. Historically, sales of this product for the period 1982 to date (1982 through summer 2003; assume 1982 was year one) have been approximated by the following data (in thousdands of units):
Y = 3.984X + 2.994
Sy = .677
r2 (R squared) = .24

(A) What would be your basic forecast of sales for 2005 (year 23)?
(B) If you wished to add the condition of 95% confidence in your 2005 forecast, what would be the range of prediction?
(C) Would you expect r2 (r squared) to have been computed using Pearson's or Spearman's mthod? Why?
(D) Suggest an alternative to time series that might improve the predictivity of the model. An r2 (r squared) of .24 isn't exactly stellar.

#### Solution Summary

We are using a linear regression model (time series) to predict sales of our "WW" brand. Historically, sales of this product for the period 1982 to date (1982 through summer 2003; assume 1982 was year one) have been approximated by the following data (in thousdands of units):
Y = 3.984X + 2.994
Sy = .677
r2 (R squared) = .24

(A) What would be your basic forecast of sales for 2005 (year 23)?
(B) If you wished to add the condition of 95% confidence in your 2005 forecast, what would be the range of prediction?
(C) Would you expect r2 (r squared) to have been computed using Pearson's or Spearman's mthod? Why?
(D) Suggest an alternative to time series that might improve the predictivity of the model. An r2 (r squared) of .24 isn't exactly stellar.

\$2.19