Reference no: EM132345731

Questions for Store24 A and B Cases -

The data required for these cases is in the file Store24AB.xls.

Summary Statistics: Definitions of the different variables in the data are provided in the case itself. Briefly discuss the summary statistics presented in Exhibit 3 in Store24A and Exhibit 2 in Store24B. Use a maximum of 2 slides for this discussion. In your team's opinion, what do the summary statistics tell us about Store24?

Doucette wants to decide whether or not to put an employee retention program in place. But first, he wants Sarah Jenkins to check whether manager tenure and crew tenure are related to store profit. Accordingly, run the three regression models per instructions given below; data for these 3 models is in the worksheet labeled Data for Case A.

Model 1: Run a full model for profit that includes both tenure and site location related variables. Tenure related variables are MTenure and CTenure. Site location related variables are population, number of competitors, street level visibility, pedestrian access, type of neighborhood, and whether a store stays open 24 hours. These variables are also defined on page 4 of the case Store24 (A).

Question - Interpret the coefficients for MTenure and CTenure in model 1. Which of the two variables contributes more to profits (i.e. which one has a greater $ impact)?

A common way of determining the relative importance of two independent variables is by assessing the incremental r-squared attributable to that variable. The following two models (Models 2 and 3) allow us determine the relative importance of manager and crew tenure in predicting profits at Store24.

Model 2: Run a model for profit with all of the independent variables from model 1 other than managerial tenure. The difference between the adjusted r-squared values for Models 1 and 2 is the incremental adjusted r-squared attributable to manager tenure.

Model 3: Run a model for profit with all of the independent variables from model 1 other than crew tenure. The difference between the adjusted r-squared values for Models 1 and 3 is the incremental adjusted r-squared attributable to crew tenure.

Question - Based on your work on models 2 and 3, whose tenure is more important in predicting profit at Store24 - manager or crew?

Model 4: Run a model with all the significant independent variables from model 1 as well as three new variables available in Case B, namely, ServQual, MgrSkill and CrewSkill. These new variables are in the worksheet labeled Data for Case B. Before running this model, see important notes 1 and 2 below.

Question - Is model 4 the best model in terms of predicting profits (relative to other models you have run up to this point)? Also, please identify variables in Model 4 that do not contribute significantly in predicting profit.

Important Note 1 before you run Model 4: In the data for part B, the variable MgrSkill has some missing values for some individuals. Regression analysis in EXCEL cannot handle missing values. So, you will have to remove individuals with missing values before running your models.

Important Note 2 before you run Model 4: When performing multiple regression analysis in EXCEL, you must have the necessary X variables in contiguous columns (i.e. adjacent columns). See the example below. First, let's assume we run a regression model for Profit with five independent variables: MTenure, CTenure, Pop, Comp and Visibility (the data below are made up; they are not from the original file). The data columns would look like that shown below. Note that all independent variables (MTenure to Visibility) are in adjacent columns with no blank columns in between.

Now suppose variables MTenure and Comp are not significant so we decide to drop them and run a new model with CTenure, Pop and Visibility, then our data would need to be rearranged like this:

MTenure and Comp need not necessarily be next to Visibility, but CTenure, Pop and Visibility must be in adjacent columns. To run multiple regression for Profit on CTenure, Pop and Visibility, we would select data in the first column for Y (Profit), and the data in the next 3 columns to include CTenure, Pop and Visibility in the model. Be sure to select the first row that contains variable names, and check labels option in the regression dialog box so that EXCEL will include variable names in the regression output.

What to submit: Submit your analysis as a power point file. For regression analysis questions, copy and paste excel regression model output (like I did in my class notes), and use a single slide to answer each question.

Attachment:- Assignment Files.rar