Week 3 LOGISTIC REGRESSION

Assignment "Credit Card Approval"

Group Task:

Assume that you work for a credit card company that currently searches for the ways to maximize profit by expanding its’ client base. The company wants to identify credit card applicants that are most likely to receive an approval and to market new credit product to clients with similar characteristics. You are asked to develop applicants’ profiles and identify groups that would be the focus of the company’s new marketing campaign.

You have retrieved information on 2,000 recent applications and their approval rates from the company’s database. You used this information to build and estimate logistic regression model. The final model included five most influential explanatory variables including indicators pertaining to the applicant’s credit worthiness, age, income, spending habits, and property ownership. The model predicts the chance of the applicant approval. Armed with the model results, you will prepare a report to your manager describing applicants’ profiles and your recommendations for the marketing campaign. You will put all technical information about particulars of logistic regression analysis into a technical note.

Group Question: Identify credit card applicants that are most likely to be approved and make recommendations for the marketing campaign targeting clients with similar characteristics

Use the following questions as a guide when formulating your response. Include detailed answers to these questions along with figures, tables and/or equations, if necessary, into the Technical Note of your report:

1. Describe you "average" customer in your data sample, which is summarized in the "Data" worksheet. The dependent variable Approval is an indicator variable taking value 1 for approved applications and 0 otherwise. There are 5 explanatory variables representing different characteristics of the applicants.

• What was the average approval rate of applications in your data sample?
• Locate quantitative variables. What was the average income, age, and credit score of applicants in the dataset? What were the highest and the lowest income in your data set? Credit score and age? Describe spending habits of the applicants in your data sample.
• Locate categorical variable, which indicates whether an applicant owes a house or rents. How many recently approved applicants owned a house?
• How would, in your opinion, these particular explanatory variables affect the chance of approval? What signs do you expect these variables to have?

1. Locate the estimated logistic regression model (worksheet "Model") and analyze the results:

1. Do the estimated coefficients have expected signs? How each variable affects the chance of the application approval? IMPORTANT: Interpret ONLY signs of the estimated coefficients. You CANNOT interpret the coefficient values as you did in traditional regression analysis.
2. Are the model coefficients significant?

1. Navigate to your individual question . At this step, each student in your group will test different hypothesis about discount means. This is an individual assignment; however, its results are important for success of your group assignment. Therefore, start working on it as early as possible. Ask for the help of your group members if needed.
2. Identify several applicants that are most likely and least likely to be approved and include their profiles along with the probability of approval into the table "Profiles of the Credit Card Applicants". You MUST come up with at least 2 profiles in addition to mine. Make up a "catchy" name for your profiles. Be creative and have fun!
3. Discuss your results and make recommendations:

• What variable(s) are the most influential?
• Based on your previous analysis, make recommendations on the marketing campaign. What customers must the targeted first?

Price: $15.46

Solution: The downloadable solution consists of 6 pages, 946 words.
Deliverable: Word Document