Regression

The sale of article X is measured on a weekly basis, The company uses various sales promoting parameters such as radio, web, store, and shop commercials etc. In the following these parameters are Called A, B, and C. The sale and parameter costs per week are shown in addendum 1.

Addendum 2 shows a number of regression analyses that all explain the sale by the parameters A, B and C - individually or combined.

1) Coment on the regression analyses set up in addendum 2 and estimate which of the models explains the sale of article x best.

In the following the management of XYZ decides to use parameter A solely to explain the sale of commodity X. See addenda 2c and 2f.

2) Set up a model showing sale as a function of parameter A. Determine and comment on the coefficients of the model.

3) Examine and comment on whether the assumptions of the model are met.

4) Assume that the costs for parameter A is 101. Calculate a 95% prediction interlval for the sale of the week in question.