STORE 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50...

Introduction

Read The Construction Department at Croq'Pain case study. 

Your submission should consist of only those exhibits that are specifically asked for. Please ensure that each exhibit is well edited and appropriately formatted. Include screen shots of stat tools inputs before creating models.

Using data on all 60 stores currently open, build a regression model that explains earnings, using the principles discussed in class.

• What is the R2of your model?  Include the regression model (the output from StatTools or other software) and mark it asExhibit A.
• Did you use transformations/interactions1?  If not, please briefly explain why not.  If you did, please create a short exhibit (one page maximum) that explains the reasoning behind including the transformation(s) interactions.  Please include the exhibit asExhibit A Extra.

Michel thinks that a good way to validate the model obtained with data from the 60 stores is to see how a similar model, using the same variables obtained from the 50 stores opened before 1994, would have performed in predicting the performance of the last ten stores opened.

• Step back one year prior to the opening of the last ten restaurants. Update the model you have developed using only the data from the first fifty stores.
• Attach your resulting regression model and mark it asExhibit B.
• Using Croq’Pain’s performance ratio target of 26%, which of the ten stores (if any) would you have opened in 1994?

Looking forward, evaluate the ten potential store locations for 1996 (see Exhibit 5). Which of these locations would you select? Base your selection on a model built on all 60 stores currently open.

Create a small exhibit that shows how you evaluate the new stores. Include the exhibit as Exhibit C. Which stores (if any) do you recommend that Croq’Pain open?

Based on what you have learned from the regression modeling in Parts A through C, what would you suggest to Croq’Pain as their new location strategy? How would you suggest that Croq’Pain use the regression models you built (or if you feel there is a reason not to base location decisions on the regression modeling, please state so and why) when scouting for and selecting new locations?

[1] This case was not written to demonstrate interactions/transformations. Please use them if you see fit, but do not include them just for the sake of including them.