Lab $1-2$ Data Analytics Questions in

Managerial Accounting

Case Summary: Each day as you work in your company$$s credit department, you

must evaluate the credit worthiness of new and existing customers. As you observe the

credit application process, you wonder if there might be an opportunity to look at data

from consumer lending to see if you can help improve your company$$s process. You

are asked to evaluate LendingClub, a U$.$S$.-$based, peer$-$to$-$peer lending company,

headquartered in San Francisco, California. LendingClub facilitates both borrowing

and lending by providing a platform for unsecured personal loans between $$1,000$ and

$$35,000.$ The loan period is for either $3$ or $5$ years. You should begin by identifying

appropriate questions and developing a hypothesis for each question. Then, using

publicly available data, you should identify data fields and values that could help

answer your questions.

Lab $1-2$ Part $1$ Identify the Questions

Your company currently collects information about your customers when they apply for credit

and evaluates the credit limit and utilization, or how much credit is currently being used, to

determine whether a new customer should be extended credit. You think there might be a way

to provide better input for that credit decision.

When you evaluate the criteria LendingClub lenders use to evaluate loan applications, you

notice that the company uses credit application data to assign a risk score to all loan applicants.

This risk score is used to help lenders determine $(1)$ whether a loan is likely to be repaid and $(2)$

what interest rate approved loans will receive. The risk score is calculated using a number of

inputs, such as employment and payment history. You have been asked to consider if there may

be better inputs to evaluate this given that the number of written$-$off accounts has increased in

the past $2$ years. Using available data, you would like to propose a model that would help create

a risk score that you could apply to your own company$$s customers.

Lab $1-2$ Part $1$ Analysis Questions $($LO $1-3,1-4)$

AQ$1.$ Use what you know about loan risk $($or search the web if you need a refresher$)$ to

identify three different questions that might influence risk. For example, if you suspect

risky customers live in a certain location, your question might be $$Where do the

customers with highest risk live?$$

AQ$2.$ For each question you identified in AQ$1,$ generate a hypothetical answer to each

question to help you identify what your expected output would be$.$ You may use some

insight or intuition or search the Internet for ideas on how to inform your hypothesis.

For example: $$Hypothesis: High$-$risk customers likely live in coastal towns.$$

AQ$3.$ Finally, identify the data that you would need to answer each of your questions. For

example, to determine customer location, you might need the city, state, and zip code.

Additionally, if you hypothesize a specific region, you$$d need to know which cities,

states, and$/$or zip codes belong to that region.

Lab $1-2$ Part $2$ Master the Data

Now that you have an idea of what questions would help influence your risk model and the types

of data that you need to collect, it is time to evaluate the specific data

that LendingClub collects using a listing of data attributes that it collects in Table $1-2$A$.$ Look

through the list of attributes and review the description of each, thinking about how these might

influence a risk score.