Author Topic: New to LC/Few Questions/Advice  (Read 4585 times)

Shawnthgreta

  • Newbie
  • *
  • Posts: 29
    • View Profile
    • Email
New to LC/Few Questions/Advice
« on: November 15, 2016, 04:12:20 PM »
I am fairly new to LC, I am starting with 5k in capital to get a feel of how this works exactly. I have read extensively on these forums, as well as blogs. I have a basic idea of how P2P lending works etc. I am just looking for a bit of advice.

1. Best tools to use with P2P lending. Right now I am considering using Peer Lending Server and NSR for filters.

2. P2P strategy. I am young and looking for an aggressive investing strategy. My initial thoughts are C-G 0 6 month inquires, 3+ years employed with no collections/public records. Any advice on this? It appears to back-test well.

3. It appears not many quality loans are listed, any advice on how to deploy 5k of capital efficiently? How long should it take to deploy? Looking at NSR notes/day, things aren't looking up.

4. Any suggested reading? A lot of what I read says diversify across 200+ notes and filter based on back-testing. Let me know if you have any other advise.

5. Does anyone know any non-paid tools to judge credit risk on notes (I would prefer not to rely on Lending Clubs underwriting 100%, since I think as a public company they face significant volume pressure)? This is especially true given the recentish changes and scandals.

6. Any other general advice or insights.

I appreciate your help in starting to learn more about P2P lending.

Thanks,
Shawn

 

Fred93

  • Hero Member
  • *****
  • Posts: 2129
    • View Profile
Re: New to LC/Few Questions/Advice
« Reply #1 on: November 15, 2016, 04:25:58 PM »
2. P2P strategy. I am young and looking for an aggressive investing strategy. My initial thoughts are C-G 0 6 month inquires, 3+ years employed with no collections/public records. Any advice on this? It appears to back-test well.

Be careful with back testing.  Suggest you do some back testing using just the 2007-2009 vintage loans, and look to see how those G loans performed during the recession.  When you back test using all history, you are heavily weighting recent years.  LC has grown very fast, roughly doubling every year, so the portfolio of all loans is heavily weighted toward recent history.  Your decision about which grades to invest in should (in my opinion) be based on what performance you can stand in the next recession.  Can't argue with your other criteria.

fliphusker

  • Sr. Member
  • ****
  • Posts: 463
    • View Profile
    • Email
Re: New to LC/Few Questions/Advice
« Reply #2 on: November 15, 2016, 05:05:16 PM »
Your on the right track by coming to the forum here, best move I did.  Lots of extremely helpful people with solid advice.  I came here in March as a complete noob, and now I am just a semi-educated noob.  :P
I spent many hours backtesting and hope that it was helpful in my initial investment in LC.  Be very careful setting up your automated stuff as I set up automated on another site and it was set up incorrectly.  So I still sit on a bunch of notes I do not want and have had defaults from them. 
I no longer invest in LC primary and have moved solely into the FOLIO market.  I know that FOLIO can be intimidating to learn, but there are good deals to be had there.  It is definitely not for everyone.
Follow the forums here and dig through them more.  But for the most part, you have to decide what is the best course of action with the amount of risk you want to tolerate. 
Personally I would not worry too much about LC underwriting.  Ya I know that some defaults are up a bit on some of their lower classes of notes.  I only use FOLIO now not because I do not have faith in LC, I do it because others do not have faith and want out. 
I think you are on the right track with your filters, might even want to snug them up even more.  When I was buying notes I had I believe 9 filters.  60k min income, no notes from CA, NV, FA or AZ, own or have a mortgage, under 20% DTI and FICO over 680.  These were just me and sure others would disagree with some of them.  But the more you filter the less notes means bigger cash drag. 
Good luck!!!

Shawnthgreta

  • Newbie
  • *
  • Posts: 29
    • View Profile
    • Email
Re: New to LC/Few Questions/Advice
« Reply #3 on: November 16, 2016, 05:17:31 PM »
Thanks for the replies. I checked out the recession (2007-2009 loans). I think F+G under performs. I modified by strategy some. Today I have gotten two loans in each drop via PLS.

Fliphusker, any reading material on Folio. I keep reading Anil posted a really good piece once, but I can't find it. Any strategies that can work there? Not looking to be super speculative. I am fairly good at analytics. Trading pure YTM work well? I would imagine you have to trade automated here to get the "good deals", any advice on manual strategies or tools to automate.

fliphusker

  • Sr. Member
  • ****
  • Posts: 463
    • View Profile
    • Email
Re: New to LC/Few Questions/Advice
« Reply #4 on: November 16, 2016, 09:09:16 PM »
Anything Anil writes is solid IMO.  If he said the Earth was flat, I might actually believe him, until he tells me he was pulling my leg.  :P 
Here is where Anil slapped me around and really made me rethink my initial strategy for FOLIO, and hence I changed it.  I was taking basically my filters from the primary market and using them on FOLIO.  Well, basically most are of zero use.  Does it really matter if someone has inquiries that are 2 years old or stable FICO and payments?
http://forum.lendacademy.com/?topic=3928.0
A bit of a YTM thread, good reading. 
http://forum.lendacademy.com/index.php/topic,2552.0.html
Let me tell you first and foremost, what I believe with FOLIO is speculation.  There is no data that supports any selling or buying on FOLIO.  I am not a financial wiz like a lot here and I am a very small fish.  I do not automate, I hand pick.  A program would have to pick up FICO changes that are recent, even though the note might still have a FICO up or flat it will not show that actually the FICO is plunging and may be in trouble. 
I have no doubt I miss out on "good deals" on notes I would normally buy, most notably ones where someone mismarks it.  But what I think of a good deal would not fit the criteria for another. 
As far as a strategy and risk....  There are endless strategies that I think can work on FOLIO.  That is where the speculation comes in.  Is a note that has 12 months of payments and an up FICO safer than a newly listed note on the primary market?  I would say for the most part yes.  But I have had notes go bankrupt out of the blue or just stopped paying like the primary market faces.  (My FOLIO ANAR is slowly recovering from 2 such notes.)  I do not buy notes at a premium so if the guy pays off his loan right after I buy it, I do not lose money as after 12 months LC charges full 1%. 
FOLIO is a crazy place and I really think 99% of all notes there are either not trying to be sold or that are listed by people who do not understand their value.  As your familiar with NSR already you can dig through the secondary market via them.  You can buy through them with no fees, until your account in 20k, I believe. 
Thanks for the replies. I checked out the recession (2007-2009 loans). I think F+G under performs. I modified by strategy some. Today I have gotten two loans in each drop via PLS.

Fliphusker, any reading material on Folio. I keep reading Anil posted a really good piece once, but I can't find it. Any strategies that can work there? Not looking to be super speculative. I am fairly good at analytics. Trading pure YTM work well? I would imagine you have to trade automated here to get the "good deals", any advice on manual strategies or tools to automate.

Shawnthgreta

  • Newbie
  • *
  • Posts: 29
    • View Profile
    • Email
Re: New to LC/Few Questions/Advice
« Reply #5 on: November 17, 2016, 02:13:28 PM »
Thanks. That was good reading. I will experiment around with Folio some. Hopefully NSR lets me weed through all the trash. I guess it would help lower time to maturity too if I invest in shorter term notes. Does only investing in clean payment history (by reviewing the payments with no late/1 max grace) up FICO notes at a discount sound like a good strategy? With 7%+ YTM.

There doesn't seem to be a lot of dicussion of Folio strategies on here. Any advice where to find more information?

Shawnthgreta

  • Newbie
  • *
  • Posts: 29
    • View Profile
    • Email
Re: New to LC/Few Questions/Advice
« Reply #6 on: November 17, 2016, 02:33:06 PM »
Quick question:Is my criteria too strict? I am only getting about 2 loans per drop. At this rate it will take 25 days to be fully deployed.

rawraw

  • Hero Member
  • *****
  • Posts: 2757
    • View Profile
Re: New to LC/Few Questions/Advice
« Reply #7 on: November 17, 2016, 02:40:33 PM »
Quick question:Is my criteria too strict? I am only getting about 2 loans per drop. At this rate it will take 25 days to be fully deployed.
Are there any criteria which aren't meaningful to your back test but eliminate a large sample of the back test universe? There is a cost benefit to strict filters. You can reduce losses but also means you reduce volumes

Sent from my SAMSUNG-SM-G935A using Tapatalk


Shawnthgreta

  • Newbie
  • *
  • Posts: 29
    • View Profile
    • Email
Re: New to LC/Few Questions/Advice
« Reply #8 on: November 17, 2016, 02:53:11 PM »
Quick question:Is my criteria too strict? I am only getting about 2 loans per drop. At this rate it will take 25 days to be fully deployed.
Are there any criteria which aren't meaningful to your back test but eliminate a large sample of the back test universe? There is a cost benefit to strict filters. You can reduce losses but also means you reduce volumes

Sent from my SAMSUNG-SM-G935A using Tapatalk

I will look into this more, I eliminated a lot of the "low trade-off" filters already though. Part of the issue might be I am using PLS' machine learning as well. I will look into playing with this some tonight. Any advice on how long is "too long" cash drag wise for deployment of initial capital?

rawraw

  • Hero Member
  • *****
  • Posts: 2757
    • View Profile
Re: New to LC/Few Questions/Advice
« Reply #9 on: November 17, 2016, 02:54:20 PM »
Quick question:Is my criteria too strict? I am only getting about 2 loans per drop. At this rate it will take 25 days to be fully deployed.
Are there any criteria which aren't meaningful to your back test but eliminate a large sample of the back test universe? There is a cost benefit to strict filters. You can reduce losses but also means you reduce volumes

Sent from my SAMSUNG-SM-G935A using Tapatalk

I will look into this more, I eliminated a lot of the "low trade-off" filters already though. Part of the issue might be I am using PLS' machine learning as well. I will look into playing with this some tonight. Any advice on how long is "too long" cash drag wise for deployment of initial capital?
Once you lend money, assume the only way you'll get it back is if they pay you. I'd much rather go slowly and get my money back compared to handing it out to everyone

Sent from my SAMSUNG-SM-G935A using Tapatalk


Shawnthgreta

  • Newbie
  • *
  • Posts: 29
    • View Profile
    • Email
Re: New to LC/Few Questions/Advice
« Reply #10 on: November 17, 2016, 03:12:07 PM »
Quick question:Is my criteria too strict? I am only getting about 2 loans per drop. At this rate it will take 25 days to be fully deployed.
Are there any criteria which aren't meaningful to your back test but eliminate a large sample of the back test universe? There is a cost benefit to strict filters. You can reduce losses but also means you reduce volumes

Sent from my SAMSUNG-SM-G935A using Tapatalk

I will look into this more, I eliminated a lot of the "low trade-off" filters already though. Part of the issue might be I am using PLS' machine learning as well. I will look into playing with this some tonight. Any advice on how long is "too long" cash drag wise for deployment of initial capital?
Once you lend money, assume the only way you'll get it back is if they pay you. I'd much rather go slowly and get my money back compared to handing it out to everyone

Sent from my SAMSUNG-SM-G935A using Tapatalk

This was my thought exactly. Anything that moves backtested ROI more than 0.3% of so, I am keeping. I guess I could do the math around cash drag, but my thought it would only be temporary, as I can set my filter as tightly as it fully reinvests returned funds. Would you recommend deploying some capital on Folio, or waiting until I know more of how all this works? I have a solid understanding of how yield/amortized notes work. Wish there were better third party tools at evaluating credit risk, but the backtesting seems to work. My current portfolio would in theory not lose in a recession (using 2007-2009).

rawraw

  • Hero Member
  • *****
  • Posts: 2757
    • View Profile
Re: New to LC/Few Questions/Advice
« Reply #11 on: November 17, 2016, 03:14:09 PM »
Quick question:Is my criteria too strict? I am only getting about 2 loans per drop. At this rate it will take 25 days to be fully deployed.
Are there any criteria which aren't meaningful to your back test but eliminate a large sample of the back test universe? There is a cost benefit to strict filters. You can reduce losses but also means you reduce volumes

Sent from my SAMSUNG-SM-G935A using Tapatalk

I will look into this more, I eliminated a lot of the "low trade-off" filters already though. Part of the issue might be I am using PLS' machine learning as well. I will look into playing with this some tonight. Any advice on how long is "too long" cash drag wise for deployment of initial capital?
Once you lend money, assume the only way you'll get it back is if they pay you. I'd much rather go slowly and get my money back compared to handing it out to everyone

Sent from my SAMSUNG-SM-G935A using Tapatalk

This was my thought exactly. Anything that moves backtested ROI more than 0.3% of so, I am keeping. I guess I could do the math around cash drag, but my thought it would only be temporary, as I can set my filter as tightly as it fully reinvests returned funds. Would you recommend deploying some capital on Folio, or waiting until I know more of how all this works? I have a solid understanding of how yield/amortized notes work. Wish there were better third party tools at evaluating credit risk, but the backtesting seems to work. My current portfolio would in theory not lose in a recession (using 2007-2009).
How are you judging that? Don't use historical rates, they change over time. Need to use current rates with expected defaults.

Second, does your screen exclude note types that didn't exist in 07 to 09?  If not, you'll be ignorantly bliss of how they'll perform until it matters

Sent from my SAMSUNG-SM-G935A using Tapatalk


Shawnthgreta

  • Newbie
  • *
  • Posts: 29
    • View Profile
    • Email
Re: New to LC/Few Questions/Advice
« Reply #12 on: November 17, 2016, 03:23:25 PM »
Quick question:Is my criteria too strict? I am only getting about 2 loans per drop. At this rate it will take 25 days to be fully deployed.
Are there any criteria which aren't meaningful to your back test but eliminate a large sample of the back test universe? There is a cost benefit to strict filters. You can reduce losses but also means you reduce volumes

Sent from my SAMSUNG-SM-G935A using Tapatalk

I will look into this more, I eliminated a lot of the "low trade-off" filters already though. Part of the issue might be I am using PLS' machine learning as well. I will look into playing with this some tonight. Any advice on how long is "too long" cash drag wise for deployment of initial capital?
Once you lend money, assume the only way you'll get it back is if they pay you. I'd much rather go slowly and get my money back compared to handing it out to everyone

Sent from my SAMSUNG-SM-G935A using Tapatalk

This was my thought exactly. Anything that moves backtested ROI more than 0.3% of so, I am keeping. I guess I could do the math around cash drag, but my thought it would only be temporary, as I can set my filter as tightly as it fully reinvests returned funds. Would you recommend deploying some capital on Folio, or waiting until I know more of how all this works? I have a solid understanding of how yield/amortized notes work. Wish there were better third party tools at evaluating credit risk, but the backtesting seems to work. My current portfolio would in theory not lose in a recession (using 2007-2009).
How are you judging that? Don't use historical rates, they change over time. Need to use current rates with expected defaults.

Second, does your screen exclude note types that didn't exist in 07 to 09?  If not, you'll be ignorantly bliss of how they'll perform until it matters

Sent from my SAMSUNG-SM-G935A using Tapatalk

Rawraw, sorry for my ignorance/naiveness. What notes didn't exist back then? I have been using 2007-2009 on my normal filter to test through recession. And then 2011-2013 for current returns (assuming the notes would have experienced the full term, therefore either defaulted or paid). Any advice on how I should change?

rawraw

  • Hero Member
  • *****
  • Posts: 2757
    • View Profile
Re: New to LC/Few Questions/Advice
« Reply #13 on: November 17, 2016, 03:40:26 PM »
Quick question:Is my criteria too strict? I am only getting about 2 loans per drop. At this rate it will take 25 days to be fully deployed.
Are there any criteria which aren't meaningful to your back test but eliminate a large sample of the back test universe? There is a cost benefit to strict filters. You can reduce losses but also means you reduce volumes

Sent from my SAMSUNG-SM-G935A using Tapatalk

I will look into this more, I eliminated a lot of the "low trade-off" filters already though. Part of the issue might be I am using PLS' machine learning as well. I will look into playing with this some tonight. Any advice on how long is "too long" cash drag wise for deployment of initial capital?
Once you lend money, assume the only way you'll get it back is if they pay you. I'd much rather go slowly and get my money back compared to handing it out to everyone

Sent from my SAMSUNG-SM-G935A using Tapatalk

This was my thought exactly. Anything that moves backtested ROI more than 0.3% of so, I am keeping. I guess I could do the math around cash drag, but my thought it would only be temporary, as I can set my filter as tightly as it fully reinvests returned funds. Would you recommend deploying some capital on Folio, or waiting until I know more of how all this works? I have a solid understanding of how yield/amortized notes work. Wish there were better third party tools at evaluating credit risk, but the backtesting seems to work. My current portfolio would in theory not lose in a recession (using 2007-2009).
How are you judging that? Don't use historical rates, they change over time. Need to use current rates with expected defaults.

Second, does your screen exclude note types that didn't exist in 07 to 09?  If not, you'll be ignorantly bliss of how they'll perform until it matters

Sent from my SAMSUNG-SM-G935A using Tapatalk

Rawraw, sorry for my ignorance/naiveness. What notes didn't exist back then? I have been using 2007-2009 on my normal filter to test through recession. And then 2011-2013 for current returns (assuming the notes would have experienced the full term, therefore either defaulted or paid). Any advice on how I should change?
Don't think 60 month notes existed. LC has also modified underwriting since then, allowing some people who previously were rejected. You'd have to Google those changes, I can't remember

Sent from my SAMSUNG-SM-G935A using Tapatalk


Shawnthgreta

  • Newbie
  • *
  • Posts: 29
    • View Profile
    • Email
Re: New to LC/Few Questions/Advice
« Reply #14 on: November 17, 2016, 04:32:36 PM »
I am only dealing with 36 month notes. I think my credit score based filter would get rid of the lower quality borrowers they took on later. Thanks for the caution though. I will Google later to make sure.