Clicky

  • Welcome to P2P Lending / NFT Lending Forum.
 

ETH.LOAN

News:

This was the original Lend Academy peer-to-peer lending forum, since forensically restored by deBanked and now reintroduced to eth.loan.

To restore access to your user account, email [email protected]. We apologize for errors you may experience during the recovery.

Main Menu
NEW LOANS:   | 804.eth 2.500 Ξ | remoraid.eth 0.299 Ξ | remoraid.eth 0.299 Ξ | ALL

wide range of FICO scores for A loans

Started by Peter, September 26, 2017, 11:00:00 PM

Previous topic - Next topic

twigster

I started converting over my portfolio to A loans since December of 2016.  At that time I had basically no A loans, now I have about 1100.  Since they are all fairly new loans (just reinvesting to buy new A loans), there haven't been that many bad loans yet, so far there are 6 in late 16+ through charged off.  I notice that there is a tendency for the bad loans to have lower initial fico scores (say 680-684) and the initial FICO range is very broad for A loans in general. 

I will note that I backtested screening on FICO in nsr and that screening on FICO has a slight negative outcome historically.

Yesterday I did a screen on existing loans to invest in with FICO greater than 750.  Most of those returned were A loans, although there was a wide range.  There were even some E and F loans in there.  I didn't see right off why the E and F loans with above a 750 FICO would be assigned that grade level.  I have attached screen shots of two of the these high FICO high interest rate loans.

I have general confusion of how lending club uses FICO in the underwriting process and why certain loans with lousy FICO could get an A grade and loans with really good FICO could get a E grade.  I would expect the assigned grade to be more correlated with FICO than it appears to be.  An insights appreciated.


Peter

They're going to hate me for using this analogy, but, you know how there's a certain amount of rat-dropping that's acceptable in hotdogs?

Suspect the same thing is going on here - kinda like mixing up a bucket of paint - you add enough of everything in the bucket to hit target percentage, after losses, and there's your bucket of A's.

Now, this works out fine if everyone gets, say, 100 A's - but when further "filtering" is applied on top of this - whether by the sophisticates or the naive - then it's possible to get a bucket full of paint that has a sufficient number of the wrong pigment flakes in it that the color is a mismatch to the desired shade.

I can write that in the parlance, if I want, but I figure it's better just to let the analogy stand and have you guys think about it for yourselves.
Publisher of the Lend Academy blog

See my returns here: http://www.lendacademy.com/returns

Fred93

Yes, there is a wide range of FICO scores present within each loan grade.  Last year I did a little analysis of this and was shocked by the spread.  Here's the chart I made.

https://forum.lendacademy.com/proxy.php?request=http%3A%2F%2Ffred93.com%2Ffbi%2FLC-FICO-vs-Grade-2016-10.png&hash=06e3776b2eeb0d3bf6d99b1b92837983" alt="" class="bbc_img" />

By the way, here's an old long thread about this:  http://forum.lendacademy.com/index.php/topic,4149.0.html" class="bbc_link" target="_blank">http://forum.lendacademy.com/index.php/topic,4149.0.html

rawraw

FICO doesn't have the income data. I suspect this is a key variable (DTI, etc) that allows low FICO in A. Not positive though




sean3.eth


NEW LOANS:   | 804.eth 2.500 Ξ | remoraid.eth 0.299 Ξ | remoraid.eth 0.299 Ξ | ALL