Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - jz451

Pages: 1 2 3 [4] 5 6 ... 8
Investors - LC / Portfolio Efficient Frontier
« on: October 30, 2016, 01:55:48 PM »
Has anyone made an efficient frontier model for their portfolio showing portfolios along the curve with the minimum variance for a given expected return? I'm having a hard time making a variance/covariance matrix because the yearly data I'm using for each grade  from 2012-2015 produces values that are essentially zero with little to no variance/covariance. My matrix for grades A-G looks like this with the filters I use for each grade on Folio: Similar results occur with primary market filters also.

   A                            B                    C                    D                     E                     F                    G
A   0.00000905   0.00002128   0.00002712   0.00002579   0.00003233   0.00002002   0.00002472
B   0.00002128   0.00006770   0.00007329   0.00006886   0.00007666   0.00003445   0.00005669
C   0.00002712   0.00007329   0.00008796   0.00008341   0.00009832   0.00004778   0.00006351
D   0.00002579   0.00006886   0.00008341   0.00007915   0.00009362   0.00004534   0.00005921
E   0.00003233   0.00007666   0.00009832   0.00009362   0.00011631     0.00006735   0.00008149
F   0.00002002   0.00003445   0.00004778   0.00004534   0.00006735   0.00007228   0.00008984
G   0.00002472   0.00005669   0.00006351   0.00005921   0.00008149   0.00008984   0.00012917

This martrix results in only two grades being invested instead of a mix of all seven grades. For example:

                             A             B             C               D            E
weights      20.68%   0.00%   0.00%   79.32%   0.00%
return terms   1.51%   0.00%   0.00%   13.49%   0.00%
target   15.0%
st dev   0.47%

I haven't tried using monthly data yet but I'de expect a similar result. I also tested a portfolio of five stocks and the model worked so something is wrong with the data being used.

General Lending Club Discussion / Re: Car refis
« on: October 29, 2016, 11:06:55 PM »
Does anyone know if these refinance loans are collateralized by the car? I'de assume yes but I can't find anywhere that says collateral is used.

Investors - LC / Re: new to investing in LC
« on: October 26, 2016, 02:49:56 PM »
What time period were you using for your data, yearly, monthly, because I'm in the process of making an Efficient Frontier of the filters I use  with yearly data and I'm having trouble with the variance/covariance table where the numbers are all essentially zero, leading to only two of the grades being given weights. 

I respectfully disagree. Because 2 loans of different grades may be less correlated than 2 loans of the same grade, there is some benefit is mixing grades. We did some analysis that one can read here:

No I don't have any program, most likely I'de use Lending Robot since NSR requires a steep $20k account to use automation. I was just saying that in the future I'll probably test between letting automation pick notes vs using my filters and looking at only FICO charts and payment history. Currently I also look at the original loan info which filters stuff out but also adds time to the process.
As my filters take out everything else, only thing I look at is the FICO and payment history as well.  As far as stricter, I would probably say different standards at least.  Last night I spent a lot of time as 3 of my filters picked up zip.  I then go through a much broader filter which is time-consuming.  Watching the debate at the same time, I enjoyed the distraction.  :P
Would you explain "test between full automation"?  I can not remember if it is you who has the program to auto buy on FOLIO or if it is another.

That's what I do as well with NSR but I must have stricter standards then you since I go through 3-5 pages just to find 1-2 notes. Ideally once I get the money to ramp up my despositing I won't have the time to manually look through notes. At that point I might test between full automation and only looking at FICO charts and payment history to save time.
Oh no, do not get me wrong, searching for notes is a chore.  But I have different saved strategies on NSR I have set up that does most of the heavy lifting for me.  Then it is just digging through a couple of dozen notes to find a few.  My account is small as well.  I have not had any pay off early but would trade my bankrupt note for a tradeoff any day.  :P  If they go IGP or later, well just the luck of the draw.  My BK was just really bad luck.
I completely agree with you about 6% being quite low end.  I will save total judgement on my strat there until a year has passed, but I think if I am under 12% for my secondary ANAR, I will be disappointed.  Long way to go to recovering from that BK though. 
FOLIO is not for everyone as a lot more work is involved finding notes. 

That was been my experience with folio as well over the past 6 months since I've started. Unlike you at some point searching for notes because a chore, even for me and my small account. And as with you prepayments are happening more often than I'de expect, out of the 96 notes I hold 6 have are paid off, but I just move on and don't let it bother me. Monitoring for grace period is another matter since I wouldn't expect many notes to go into grace period for my notes which are over 12 months old, but they are the lower grades (F-G)which I'm trying to move away from.

Higher returns are easily achievable, 6% XIRR is a bad month for me (only happened when I deposited more money after my intial deposit) unlike what almst791 is expecting, so even 10% is achievable for most people with proper filtering.

Another aspect to realize which is relation to my one bad month is that if notes are bought on Folio with large deposits compared to account value that returns will temporarily be stunted due to accrued interest and premiums being removed unlike buying fresh notes.
6% on FOLIO is very achievable in my opinion.  I am strictly now a FOLIO buyer.  I do list most of my non-FOLIO notes there but not really priced to sell with nice sized markups. 
Here are a few things about FOLIO
Buying on FOLIO is all about what you will deal with for risk tolerance.  Or what you are willing to pay for a markup.  There are realistically hundreds of different ways you can run a strategy on FOLIO.  Determining your risk is where you start to whittle down your risk tolerance.  I will tell you this though, I honestly would say that 99% of notes on FOLIO are priced incorrectly.  This is due to the fact that FOLIO is one crazy place and most do not understand it.  I have been in the FOLIO market only since May and can easily say I do not understand how notes are priced most of the time. 
Two things that are important to look for on FOLIO, IMO, is FICO score and payment history.  For me, recently down FICO or recently late are red flags and off the table.  No doubt there are people who go after these notes with their deep discounts.  The other factor for me is the YTM has to be a certain amount.  There are other factors that are part of my secret sauce.  :)
I use Nickelsteamroller to dig for my FOLIO notes.  The FOLIO site is an absolute joke and NSR gives great details and notes can be bought from their site.  Here is what I mean from filtering on NSR of notes.  440k notes listed there have not been late.  Of those 357k have a markup to them.  Of that 89k have a YTM over 15%.  Of that around 50k have a FICO of up or flat.  So lets say you want some maturing on your notes of at least 12 months that brings us to 15k notes.
I dislike paying less of a YTM than what is on the note and only like to grab bargains.  So I take all above and look for discounts.  With the above filter which leaves 3,600 notes.  FOLIO has over 500k notes listed and with only some basic filters that drops to 18k. 
The problem with FOLIO is that they do not release any numbers on how many notes they sell what grade or term.  I can only tell you that the notes I generally shop for are not easy to find as I am quite picky. 
The biggest fear I have about FOLIO notes after the first year is not so much default, but they will pay off early.  When they do that you get hit with the full 1% fee.  Especially if you are paying a markup on a note you could actually lose money on it.  Even seasoned notes can go belly up out of the blue.  My ANAR for FOLIO is still battling back from a $50 note that went bankrupt with zero issues with the note before I bought it, after 1 payment. 
FOLIO is not like LC main and you have a huge variety to choose from.
$2,500 is not too small of an amount especially in FOLIO.  As you will not be paying full face value of a note, most likely, you will diversify faster. 
Are you getting above 6% now in the market?

Broad stroke of what FOLIO actually is
Here is an old but good article.
A bit old but still valid.

I know others use automated tools to buy on FOLIO but I never will.  A note may have an up score but has dropped 60 points in the past two months.  Also FOLIO makes mistakes and will list notes that they say have never been late, but have been in IGP or worse at some point. 

FOLIO is frustrating but I also think I can get safer cheaper buys than newly issued notes.  Which you can find there too.

Investors - LC / Re: Marcus by Goldman Sachs
« on: October 15, 2016, 09:10:11 PM »
Nothing is going to give you a yearly income that's in that 7-10% range. You're pushing it with high yield corporate bonds that get can return 4-6%. You're safe bet is to simply diversify between different assets. And if you're getting only 5% from LC then your filters must not be strict or you're investing too much into the lower grades because with some simple filtering you should be yielding at least 7-9%. 
Thanks for your thoughts. I'm beginning to dislike LC because, as mentioned earlier, they seem to not care about the non-payers.
This is starting to be too much work for me. If I don't "hedge and trim" quickly enough any note that starts to get in grace period, 30-days period and so on, it rapidly turns into default.

Is there a way to make 7-10% yearly income in a relatively safe/guaranteed manner?
What about these real estate-centric lending business?

Are they riskier than P2P lending?
I have started the LC with 200 notes and a promising 9-11% yearly interest projections (have only 36-month loans) and good mix of C4, C5, down to E and F. Now, the projection got down to a measly 5%. Not worth anymore the work. Had to self-manage/trim many notes and sell them via folioN.

Thanks again.

Investors - LC / Re: LC raises rates 10/2016 and updates loss forecast #s
« on: October 15, 2016, 09:01:37 PM »
I wouldn't be too worried about looking at that chart because you can see that currently the curve is on the bottom end where the slope is not as steep as the big spikes where it shows a recession has occured.

Having vintage data is not essential.  Could start with SOME data from somewhere outside marketplace lenders that shows some aggregate delinquency rates are going up.  So far all data I have says they're going down elsewhere.  Once we can see some data that is moving up, like the LC and Prosper data, then we can begin to build a story that something broader than marketplace lending is happening.

Agreed that there must be some performance data from some securitizations somewhere that might help, but I know zero about securitizations, so have no idea where to look.

In the name of full disclosure, I did find one Fed delinquency data series that started moving up mid-2015.  That's the "commercial and industrial loans" series.

I've been ignoring data that isn't consumer installment or credit card, but this series has been moving up for several quarters, so its trend is real.  There must be a reason the C&I loan delinquencies are moving up, and its possible this is part of the picture.  On the other hand, in the 2002 recession, C&I loan delinquencies went up significantly, but consumer installment & credit card never did, so it is quite possible for C&I to go up on its own.  Every cycle is different.

Investors - LC / Re: Marcus by Goldman Sachs
« on: October 15, 2016, 08:40:12 PM »
There is none, simply another tool for Goldman to make money.
Stupid question: where is the investor aspect of the Goldman's Marcus solution?
I guess there is none? I'd like very much to be on the lending/investment side with them. Starting to HATE lendingclub (will be pulling all of my money off them). LC simply doesn't care about the collections / non-payment aspect at all, and you are left alone in the dark. Boo!

Since I haven't had my account open long enough to receive a statement with tax info I'm not even sure what it details, but I'm keeping track either way until I get to that point next year. And yes if LC provides a significantly difference I'll provide my own info with documentation.

In my view it's easy to gather the information off of the traded notes excel spreadsheet on the notes tab on LC. All the info is there any easy to arrange, sorting by transaction date and copy/paste the data where you left off. I'm using a table for my spreadsheet so the calculations automatically update for new entries.

That makes sense and seems somewhat manageable on a very small scale.  But then what will you do on your taxes....alter the numbers that LC sends you and make a note as to why?

Seems this would become exponentially more unmanageable as account size grows and if you used automatic buy/sell in Folio.  Especially when notes purchased at other than par value later default or are then sold again on Folio.  Unless someone has created a tool that can import all transactions and somehow automate all of the calculations and adjustments necessary.

Sent from my iPhone using Tapatalk

The way I am doing it now, even with such a small account ($1.5k) I made a spreadsheet where I calculate the gain/loss at the time of purchase of notes, straightline amortization that over the period of time I have held the notes, and calculate the gain/loss at the sale of any notes, and then sum up the total of all notes of amortization once the notes are sold or paid off. Obviously it is easier for me since I only have a few notes that I buy/sell each month.

Why can't the borrows take the initiative and apply to refinance? As an investor I wouldn't want that to be built in, but if they want to pay me off and have other people invest in the refinanced loan then go ahead and I will move on to another loan instead of having my money tied up even longer.

Investors - LC / Re: What is your Portfolio Analysis Tool of choice?
« on: August 04, 2016, 11:59:33 PM »
As far as I know all you need is to put in your payment information and you get to input as many criteria as you want instead of the limit of 3.

Wrote my own.  Also use NSR.  Shame that reserve the more advanced backtesting features for paying customers now.

Investors - LC / Re: What is your Portfolio Analysis Tool of choice?
« on: August 04, 2016, 09:16:34 PM »
I've been using NSR platform for my backtesting and the portfolio analysis lets you analyze all of the note criteria. Otherwise I use my own spreadsheet w/ returns, graphs, and notes that have defaulted.
As an LC & Prosper investor, I am curious as to your choice of tool/service for portfolio analysis, and why you chose what you did.

Me: Over the last 3 years I have invested in nearly 10k notes, and am always interested in finding trends amongst those notes that may help me tune my note picking strategy.

Prosper's utter lack of any analysis is maddening to me. I find LC a bit better, but not nearly what I am looking for. I use Lending Robot for investing (love their service, love their staff), but analysis & reporting is not their focus.

So far, I have created some spreadsheets, done some basic development work on a custom XAMPP based system, and used a trial account on Interest Radar to try to determine what the best option would be. Spreadsheets work, but are basic. I'm a techie and love XAMPP, but don't have the cycles to build a system from scratch. IR definitely held some promise, and so I am definitely leaning towards a 'service' like solution.

I'd love to know what tool or service people are using (and like) to do portfolio analysis & reporting. I know there is a lot out there, and I'm not sure I want to (or need to) build my own. So what do you peeps use?

Pages: 1 2 3 [4] 5 6 ... 8