Post a Comment On: Transplant Athlete

Update: Now that Lending Club has raised their loan limit, these numbers have all changed. So, please don't rely on these results. You can follow the methodology to perform your own analysis on the new data. If you're looking for the highest yields at the lowest risk on Lending Club, you really have to work at it. Let's say you have a hunch that homeowners are the safest bet on LC, how do you go about checking that? You look at the data. Go to Lending Club's Statistics Page and then ignore all the fancy charts you see. Click on Download data. Choose to "Download Loan Data" (not In-Funding or Declined).It's kind of a large file, because it contains every loan that's ever been initiated on LC. That means, Prepaid, Paid, current, not so current, defaults, charge-offs, and removed. When a loan has been in default for a long period and LC doesn't think it will ever be paid, it is "Charged-off". As far as I can tell, Loans that are "Removed" were listed, may have received some funding, and then were "Removed" prior to funding. This may have been at the Borrowers request or at Lending Clubs. What can we learn from this data? A true Math Geek could do a statistical analysis and tell which factors (income, location, home ownership, DTI, etc) CORRELATE to better loan performance. I just barely passed Statistics at business school (and that was 6 years ago), so I'll give what observations I can. First of all, there are almost 26,000 records in this file. So, I sort the data by Status, and then by Loan Grade. This allows me to see all the Charged-offs together, all the defaults, all the paid, etc. Let's get back to our guess that homeowners are better at paying off their loans. The computer doesn't understand "RENT" and "MORTGAGE", so I change those into numerical values using Find and Replace. Rent=0, Mortgage=1. When you crunch the numbers, here's what you get: Home Ownership Percentages By Loan Status Charged Off48.7% Default50% Paid50.6% Issued53.1% Late (16-30)45.8% Late (30+)46.4% I think that busts our hypothesis that homeowners are more likely to pay their loans. What's our next hunch to check? Low DTI = High Performance? Low Loan Amount = High Performance? Monthly Income? Loan Purpose? Open Credit Lines? Total Credit Lines? Revolving Credit Balance? Revolving Line Utilization?