12 ms·
A Dive Into The Lending Club Data
- Charlesmigli 11y agoNice post. I see that you used dc.js, could you share some code ?
- ClementM 11y agoThe code to process the lending club data is done in python. I got the full history of all payments made on the platform and pre-process it to have something that's light enough to be explored with a good user experience. For the viz', yes I used DC.js, I can open source the .js if you guys want it.
- Amorymeltzer 11y agoI'd imagine we'd all appreciate that!
- ClementM 11y agoI'll update the post and put a link to a github
- edelans 11y agoHi Clement, very neat dataviz ! How did you get the full history of all payments ? Is it available in the open ? What is 100mdeep btw ? Just a blogging site you are bootstraping or any plans to make a living out of it ? (We met a few months ago @ ToucanToco...)
- ClementM 11y agoI did this work for myself at first as I invested some money on the Lending Club. Plans will depend on feedbacks I receive. If it generates enough interest I will develop the project into something bigger.
- ClementM 11y ago100m is the depth human can free dive to. Without any aid. Without any fins. https://en.wikipedia.org/wiki/Constant_weight_apnea https://en.wikipedia.org/wiki/Constant_weight_apnea Tribute to people who go deep into things ;).
- edelans 11y agoNice image! I'm a big fan of Guillaume Nery, the french free diver (4 word records). Actually Constant weight divers are allowed to use fins (monofins most of the time), although they use it only to get to ~30m deep, after that point, the pressure is such that the volume of the body decreases and Archimedes thrust is no longer sufficient to compensate for the weight... so the diver can descend without any mouvement... It's very impressive to see (http://www.liveleak.com/view?i=d4a_1244462128 http://www.liveleak.com/view?i=d4a_1244462128). Funny to read about this a few days after Guillaume Nery's accident: a line was set to the wrong depth causing the diverto dive to -139m/456ft instead of -129m/-423ft that he had announced the night before...
- ClementM 11y agoFunny you mention ToucanToco. They helped me in debugging the viz.
- edelans 11y agoThanks for the replies, what about the source of the data ?
- rocketcity 11y agoThis is great. I have been using LendingRobot lately. I will have to rethink some of my strategies based on what I see in this article.
- existencebox 11y agoFor my own personal interest, can you speak at all on your experience using LendingRobot? (successes/misses, use duration, gains, etc)
- gtremper 11y agoI've been using it for about a month and its one of the better LendingClub auto-investors out there. By far the best interface, but the fee is a little steep, though the fee only takes effect after investing 10k though them. I've rolled my own auto-investor in the past and used the one on NSRPlatform(free for accounts <20k). I'm planning switch over to LendingRobot 100% when I can't use NSRPlatform for free anymore. Here's my shameless referral for lending robot https://www.lendingrobot.com/ref/YehMn307/ https://www.lendingrobot.com/ref/YehMn307/ (you get 10k free, I get 5k more free)
- existencebox 11y agoThanks! Any recommendations for someone looking to get into this space, especially in terms of things you had wish you knew starting off? (I'm looking at this as a "long shot" diversification of a portion my savings, allocating a comparatively very small amount and going entirely hands off)
- gtremper 11y agoI have all my LendingClub funds in a Roth IRA so I don't have to deal with any of the tax loss/gains stuff. I've heard the tax accounting can get tricky in normal accounts, so I highly recommend putting your LendingClub funds in a tax-advantaged account. As for filtering and stuff, you can you do your own underwriting(loan risk analysis) with the data lendingclub provides[1], or use https://www.nsrplatform.com https://www.nsrplatform.com, which has a nice GUI tool to explore the data with you're own filters. LendingClub has a JSON api, so you can an order executer for yourself. (Here's the remnants of the one I was working on https://github.com/gtremper/LoanInvestor https://github.com/gtremper/LoanInvestor. P2P-Picks was a 3rd party underwriter that isn't available anymore). I've noticed that the D and E loans tend to be the best balance of risk and return Also, be aware that you'll need to continuously buy new notes as payments come in to your account, otherwise you'll build up cash rather quickly. That's why these auto-investing services are so useful. Its best to buy only $25(the minimum) per loan so you can spread your risk among as many notes as possible. [1] https://www.lendingclub.com/info/download-data.action https://www.lendingclub.com/info/download-data.action
- Amorymeltzer 11y agoThe employment length is really bugging me. I've always selected people with a few years at their current job, leaning towards higher, because it feels safe, but this says that <2 years of experience is better than longer! I wonder if they're "newer" so more likely to stay around and not be pushed out, or if the rates are much higher compared to a marginal increase in risk. I'm leaning toward the latter. It looks like income has the same effect for home loans; <50k has a much higher return simply because they get a huge rating hit. My other big hit is 3 versus 5-year terms. Anyone here care to comment? I like the 36 months because it feels more liquid and when I started I wasn't sure LendingClub was going to be around for a decade or more. Beginning to think I should reconsider that stance.
- ClementM 11y agoYes, exactly for all the criteria it is a matter of "the rates compared to marginal increase in risk". Look at the 'A' grade. They're nice and safe intuitively, but in my opinion they're not a really good investment.
- ClementM 11y agoAlso, it's pretty instructive to look at the Lending Club grading algorithm in details. They made it public at some point. Now they are a little less transparent about it. But some details can be found in their SEC prospectus. I can link that up as well if you guys want.
- jlittel 11y agoI'd love to see any info. Their current offering document is here[1], but I don't see much mention of specifics. There's some detail on mapping to grades on p42 of the Aug 22 doc, as well as interest rates charged for each risk category. [1]https://www.lendingclub.com/info/prospectus.action https://www.lendingclub.com/info/prospectus.action
- ClementM 11y agoI have some details somewhere on my drive. PM me and we can talk
- cryoshon 11y agoHm, this has sparked my interest in lending via Lending Club. I'll check this out for sure.
- rgn216 11y agoVery useful tool. Gives valuable insight on how to select filters in portfolio construction. If "the Pearl" was a existing product, I would definitely invest in it.
- ClementM 11y agoLook for the Pearl! you can definitely get 8% interest over the long run with a good liquidity on your cash deployed. To me, it's totally worth it for money that you don't need in the really short term
- fbenezit 11y agoThanks Clement for this beautifully simple dc.js dataviz. How long did you play with it before finding the pearl? Do you think there are yet other pearls to find in your tool?
- ClementM 11y agoThere are many pearls to find. It depends on your 'set of preferences'. You want return? you want low-risk ? you care to deploy a lot of money, or not that much? But I'd say that for starter, anything in the 8%-9% range is a very good deal these days in this environment
- radmuzom 11y agoI build statistical models for banks which help assess the risk of a loan. Effectively, my models will get converted into the grades (A, B, C, D, etc.) mentioned in the article. The strategies (second chance, family guy, safe haven) are generally consistent with experiences from the portfolios of most financial institutions. However, I am skeptical (prove me wrong) of the statement in the article - "Lenders get a return on their investment that is typically much better than traditional Certificate of Deposit or Saving Accounts". In finance terms, I will be surprised if they have a higher RAROC [1] as compared to large banks. If they really do, then congratulations (you will put banks out of business in a few years)?? [1] https://en.wikipedia.org/wiki/Risk-adjusted_return_on_capital https://en.wikipedia.org/wiki/Risk-adjusted_return_on_capita...
- ClementM 11y agoGlad that my findings match yours ! I think it does beat a C.D. from a risk/return perspective. It's probably higher risk than a C.D. but returns largely compensate for it I believe. To me this is possible because the Lending Club is desintermediating a business that was traditionnally 'high margin'.
- dragonwriter 11y ago> In finance terms, I will be surprised if they have a higher RAROC [1] as compared to large banks. If they really do, then congratulations (you will put banks out of business in a few years)?? Or, more likely, just drive down bank profit margins.
- guelo 11y agoI don't know much about the banking industry but savings accounts and CDs have always felt scammy to me. They're marketed to the rubes that have no idea what they're doing so they can get away with not being competitive with other financial instruments.
- jklein11 11y agoThe reason why savings accounts and CDs have such a low interest rate is because they are far less risky than other financial instruments. If you compare them to other assets with a similar risk profile and payment structure, such as short dated treasury notes and annuities, you will find that the rates are at least competitive. Disclaimer: My response has a US bias.
- akg_67 11y agoI operate an online crowd-lending analytics and automation platform PeerCube https:/www.peercube.com. I have been analyzing both Lending Club and Prosper data for my institutional clients for almost 4 years now. While OP made a good first attempt on analyzing the data, the analysis suffers from two major shortcomings that I normally see from people getting started with data analysis. 1. Domain Knowledge: Novice analyst tend to put the data in a blender and see what comes out first instead of building some preliminary knowledge and intuition about the domain. This is quite evident in OP's analysis and finding about annual income. A person familiar with domain will ask the question "Why would a borrower with high annual income will borrow a small amount loan at high interest rate?" This right away will raise flags about risks of lending to such borrowers. OP will benefit by reading some of the publications (books, research) on credit scoring and modeling before deep diving into analyzing Lending Club data. 2. Data Exploration: Not spending enough time exploring the data can lead to erroneous conclusion like The second chance strategy. When did Lending Club start issuing loans to borrowers with delinquencies and public records has a big impact on returns as newer loans are not aged enough to have sufficient defaults. > Watch for your average return (expected return), consistency of returns through time (risk), while making sure there is enough supply (liquidity) on the platform to deploy your strategy. Time is not Risk. You need to find a proper measure for risk. Also consider negative kurtosis and frequent low positive returns but a few high negative returns nature of return distribution. > I considered that investors deploy and re-invest their money continuously on the platform and therefore own a portfolio with different ‘vintages’ of loans. The ROI that are computed reflect this, as they are average returns across vintages. Re-consider this argument of "average return across vintages" being representative of investor returns. Tip: look at loan volume across vintages as well as typical re-investment pattern of a typical investor. > Please also note than due to the low issuance volume in the early days of the platform, the returns computed for the pre-2010 period are much less reliable than the post-2010 returns. Please don't do this. The data between 2006 and 2010 is the most valuable due to the business cycle we were in at that time. The data since 2010 tells nothing about how loans might perform in the future when business cycle is not as good it has been in last few years. OP will really benefit from re-evaluating his finings with critical eyes. I will suggest gaining some domain knowledge, spending lot of time on just exploring the data before start drawing definite conclusions, focusing on distributions, correlations and statistical significance.
- mikeskim 11y agocan you fully automate data driven investment on these platforms?
- ClementM 11y agoYou can fully automate investment 'filters' i.e. filter the notes you're willing to invest or not.