Author Topic: New note selection model & website. Would love feedback  (Read 38621 times)

PeerSocialLending

  • Jr. Member
  • **
  • Posts: 79
    • View Profile
    • Email
Re: New note selection model & website. Would love feedback
« Reply #30 on: November 02, 2014, 11:55:00 AM »

I see what you are saying.  Maybe we should add a 3rd percentile then, similar age AND similar weighted average interest rate.  That data is in the JSON so it is just a matter of pulling it out and calculating it.

I think that would be a great feature.  You could even add the filters like Lending Club has so we can view our percentile based on factors we choose :)
-Ryan
p2p lending blogger @ www.peersociallending.com

turing

  • Full Member
  • ***
  • Posts: 222
    • View Profile
    • PeerToPeerQuant.com
    • Email
Re: New note selection model & website. Would love feedback
« Reply #31 on: November 03, 2014, 11:22:29 AM »

I see what you are saying.  Maybe we should add a 3rd percentile then, similar age AND similar weighted average interest rate.  That data is in the JSON so it is just a matter of pulling it out and calculating it.

I think that would be a great feature.  You could even add the filters like Lending Club has so we can view our percentile based on factors we choose :)

This is a good suggestion too.  Thanks Ryan.  I have added these to our list of improvements for Percentile tool and hopefully we can add some of these in the near future.

Zach

  • Administrator
  • Hero Member
  • *****
  • Posts: 622
    • View Profile
    • Email
Re: Re: New note selection model & website. Would love feedback
« Reply #32 on: November 03, 2014, 10:15:25 PM »
BTW you may want to request a subforum for your product.
All set :D

turing

  • Full Member
  • ***
  • Posts: 222
    • View Profile
    • PeerToPeerQuant.com
    • Email
Re: New note selection model & website. Would love feedback
« Reply #33 on: November 05, 2014, 12:49:41 PM »
Thanks Zach.

lascott

  • Hero Member
  • *****
  • Posts: 1435
    • View Profile
    • Appreciate my post and want to try LendingRobot? URL below
Re: New note selection model & website. Would love feedback
« Reply #34 on: December 24, 2014, 04:36:59 PM »
Ran across this old blog while searching for something else.  Thought those following this thread my find it interesting.

http://blog.dmpatierno.com/post/3161338411/lending-club-genetic-algorithm

Quote
February 07, 2011 1 note
Using genetic algorithms to maximize Lending Club performance.

Iíve had an account with Lending Club for over 3 years. In case youíre unfamiliar, Lending Club offers borrowers lower interest rates than a credit card company and provides individual lenders (like me) higher returns than a bank. They take about a 1% cut, but handle all the details like automatic withdrawals, payment-plan negotiation, and collections. Itís a win-win-win situation. They boast an (impressive) average 9.68% APY for investors, but I wanted to see if I could beat that without spending hours picking through individual loans every day.

So I wrote some code.

<snip>
Tools I use: (main) BlueVestment: https://www.bluevestment.com/app/pricing + https://www.interestradar.com/ , (others) Lending Robot referral link: https://www.lendingrobot.com/ref/scott473/  & Peercube referral code: DFVA9Y

turing

  • Full Member
  • ***
  • Posts: 222
    • View Profile
    • PeerToPeerQuant.com
    • Email
Re: New note selection model & website. Would love feedback
« Reply #35 on: December 24, 2014, 05:20:20 PM »
Good find on that link.  I'm going to reach out to him to talk shop a bit to hear more about the methods he used and how he combated overfitting.

The biggest hurdle we found with genetic algorithms is making sure you build in methods to combat overfitting (regularization, cross sample testing, early stopping, etc).  Other wise the model starts predicting noise instead of predicting the underlying trend.

Thanks for sharing that link.

rawraw

  • Hero Member
  • *****
  • Posts: 2793
    • View Profile
Re: New note selection model & website. Would love feedback
« Reply #36 on: December 25, 2014, 04:59:52 AM »
Pretty sure Bryce went head to head with that guy in a backtest white paper.  He took the posted strategy and backtested it compared to his

turing

  • Full Member
  • ***
  • Posts: 222
    • View Profile
    • PeerToPeerQuant.com
    • Email
Re: New note selection model & website. Would love feedback
« Reply #37 on: December 26, 2014, 10:41:37 AM »
Pretty sure Bryce went head to head with that guy in a backtest white paper.  He took the posted strategy and backtested it compared to his

That would be a very interesting read.  Do you have the link for it by chance?  I don't see it with a quick google search?

rawraw

  • Hero Member
  • *****
  • Posts: 2793
    • View Profile
Re: New note selection model & website. Would love feedback
« Reply #38 on: December 27, 2014, 07:21:14 AM »
Pretty sure Bryce went head to head with that guy in a backtest white paper.  He took the posted strategy and backtested it compared to his

That would be a very interesting read.  Do you have the link for it by chance?  I don't see it with a quick google search?
I tried finding it recently to show a friend.  But I think Bryce took it down from the website for whatever reason.

brycemason

  • Hero Member
  • *****
  • Posts: 801
    • View Profile
    • P2P-Picks.com
    • Email
Re: New note selection model & website. Would love feedback
« Reply #39 on: December 28, 2014, 10:39:06 PM »
I included that genetic selection criteria in my portfolio comparison white paper. However, it wasn't a terribly fair inclusion as it didn't pick a ton of notes and the backtest period overlapped somewhat with the original author's testing period. I wound up taking it out in a later version. I also thought the best comparison for a product like mine was something that would pick at least 10% of 36 month loans (in the appropriate grades). The method of comparison was good, though, and I haven't seen anything like that since.

turing

  • Full Member
  • ***
  • Posts: 222
    • View Profile
    • PeerToPeerQuant.com
    • Email
Re: New note selection model & website. Would love feedback
« Reply #40 on: December 29, 2014, 01:22:29 PM »
I included that genetic selection criteria in my portfolio comparison white paper. However, it wasn't a terribly fair inclusion as it didn't pick a ton of notes and the backtest period overlapped somewhat with the original author's testing period. I wound up taking it out in a later version. I also thought the best comparison for a product like mine was something that would pick at least 10% of 36 month loans (in the appropriate grades). The method of comparison was good, though, and I haven't seen anything like that since.

Thanks for letting us know Bryce.  Do you still publish that white paper without the genetic algorithm comparison? 

lascott

  • Hero Member
  • *****
  • Posts: 1435
    • View Profile
    • Appreciate my post and want to try LendingRobot? URL below
Re: New note selection model & website. Would love feedback
« Reply #41 on: December 29, 2014, 02:33:32 PM »
I included that genetic selection criteria in my portfolio comparison white paper. However, it wasn't a terribly fair inclusion as it didn't pick a ton of notes and the backtest period overlapped somewhat with the original author's testing period. I wound up taking it out in a later version. I also thought the best comparison for a product like mine was something that would pick at least 10% of 36 month loans (in the appropriate grades). The method of comparison was good, though, and I haven't seen anything like that since.

Thanks for letting us know Bryce.  Do you still publish that white paper without the genetic algorithm comparison?
Nutshell:
Comparison of Different P2P Lending Investment Strategies
by PETER RENTON on MAY 8, 2013
http://www.lendacademy.com/comparison-of-different-p2p-lending-investment-strategies/
Tools I use: (main) BlueVestment: https://www.bluevestment.com/app/pricing + https://www.interestradar.com/ , (others) Lending Robot referral link: https://www.lendingrobot.com/ref/scott473/  & Peercube referral code: DFVA9Y

turing

  • Full Member
  • ***
  • Posts: 222
    • View Profile
    • PeerToPeerQuant.com
    • Email
Re: New note selection model & website. Would love feedback
« Reply #42 on: December 29, 2014, 06:01:48 PM »
Thank you lascott.  Some great reading material for me to go over. 

brycemason

  • Hero Member
  • *****
  • Posts: 801
    • View Profile
    • P2P-Picks.com
    • Email
Re: New note selection model & website. Would love feedback
« Reply #43 on: December 31, 2014, 03:10:03 AM »
I'll send you a copy, just shoot me an email.

turing

  • Full Member
  • ***
  • Posts: 222
    • View Profile
    • PeerToPeerQuant.com
    • Email
Re: New note selection model & website. Would love feedback
« Reply #44 on: December 31, 2014, 11:03:10 AM »
I'll send you a copy, just shoot me an email.

Email sent.  Thank you