Skip to content

use Python’s data visualization tools to systematically explore a selected dataset for its properties and relationships between variables. Then, create a presentation that communicates the findings to others.

Notifications You must be signed in to change notification settings

xxiMiaxx/Communicate-Data-Findings

Repository files navigation

Loan Data From Prosper

by Lamia Alsalloom

Dataset - Prosper Loans

Prosper Marketplace is America's first peer-to-peer lending marketplace, with over $7 billion in funded loans. Borrowers request personal loans on Prosper and investors (individual or institutional) can fund anywhere from $2,000 to $40,000 per loan request. learn more

This data set contains 113,937 loans with 81 variables on each loan, including loan amount, borrower rate (or interest rate), current loan status, borrower income, and many others.

Summary of Findings

Most of the loans are either Current or already Completed in regards of loan status.

As for the loan duration 87778 loans have a length of 36 months, then 24545 loans have a length of 60 months and 1614 loans have a length of 12 months.

A huge portion of the loans appear to be Debt Consolidation loans.

As for the income range we found the following: We can break people into groups in here, the first group is of people who have an income that ranges from $25,000-49,999 this group have the biggest number of listings, meaning they take loans and borrow money more than the rest.

The second group have income that ranges from $50,000-74,999 those are not very diffirent from the first group.

The third group is for the rich folks who have an income range of $100,000+, they still borrow money but not as much as the first two groups.

The fourth group is for people who are almost rich, with an income range of $75,000-99,999, their listings are similir to the rich folk

The last group is for people who have little to no income, or the data was not displayed, as expected their listings are very few cause how else are they gonna pay back the loan.

then moving to the bivariant exploration we observed the change of the borrowers rate over the years

we found that rate of borrowers interest spiked up from 2005 to 2006, after that the rate dropped from 2006 to 2007, then the rate constantly kept increasing for the next few years until it reached its highest in 2011 which after that it kept decreasing constantly.

The Effect of CreditGrade over the LoanStatus: most of the loans have a C credit grade while B and D are similar in the number of loans.

We found most of the loans are Completed regardless of the CreditGrade, this is an obvious trend.

How the AvailableBankcardCredit effects that BorrowerRate: the more credit one has at the bank the less loans needed and therefor less borrower rate at hand.

The correlation between LoanStatus and BorrowerAPR: completed loans have lower Borrowers APR compared to the distribution of defaulted and cahrgedoff loans.

Relationship between the BorrowerAPR and CreditGrade: uncompleted loans have a higher mean APR than those of completed loans, across most grades.

Completion rates are high for listings with better Credit Grades.

Do Homeowners and Non-Homeowners have a variant BorrowerRate over the years: the borrowers rate has increased over some years in case of non-homeowners then it kept decreasing in 2010. And when we look at homeowners it kept a steady pace until after 2009 it kept increasing until 2011 then started to decrease.

Key Insights for Presentation

completed loans have lower Borrowers APR compared to the distribution of defaulted and cahrgedoff loans.

uncompleted loans have a higher mean APR than those of completed loans, across most grades.

Completion rates are high for listings with better Credit Grades.

Do Homeowners and Non-Homeowners have a variant BorrowerRate over the years: the borrowers rate has increased over some years in case of non-homeowners then it kept decreasing in 2010. And when we look at homeowners it kept a steady pace until after 2009 it kept increasing until 2011 then started to decrease.

About

use Python’s data visualization tools to systematically explore a selected dataset for its properties and relationships between variables. Then, create a presentation that communicates the findings to others.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published