pikestaff
Member of DD Central
Posts: 2,187 
Likes: 1,546
|
Post by pikestaff on Feb 13, 2015 23:33:58 GMT
I am utterly baffled by the GEIA. I parked some money there on 4 Feb, with a view to taking it out again when one of the upcoming loans draws down. All but a tiny amount got invested PDQ. So far so good. A couple of days ago I set it to withdraw repayments and interest, at which point it started to SELL my investments for NO reason. My funds actually invested and earning were £1,700, and are now down to £1,281.89. And because sales are not repayments, the money is just sitting there doing nothing. I will now turn off "withdraw repayments and interest" and see if it reinvests my funds... Where did you get the idea that GEIA was a cash account? Its a long term investment and depends on other lenders being available to buy your investments so you can exit early. A software change was made earlier this year to unstick some investments which did not appear as available to other, but other that that you have to wait for the randomiser to select your investments for a willing lender when there is a large queue. I know it's not a cash account and that it may take time to buy and sell. That's not my point. I was getting sales without any concurrent purchases (the "churning" that many have complained of), despite having made NO instruction to sell. That's just wrong. The speed of sales is supposed to be much faster following changes to the algorithm, and that's borne out by my experience this afternoon/evening. The loans I was waiting for appeared on the market this afternoon, and I changed the target on my green account to nil as soon as I got the email. It's sold about 2/3 of my balance so far, which I've used to buy the loans I want. I've topped the loans up to my targets with cash freed up by selling selected holdings in my MLIA (ones it's easy to buy back). When the remaining sales from my green account come through I will use the proceeds to restore the latter holdings to their previous amounts. Hopefully that won't take more than a few days, but we shall see... Edit chris, my post crossed with yours. I don't think that diversification can be the explanation. I had £1,700 just about fully invested. When I asked to withdraw repayments and interest it started making sales for no reason at all. I can see that a repayment on one loan might trigger equal sales on all the others, although this is not desirable IMO (see below*). However, any such sales should be for small [net] amounts and this is not what was happening. It sold a substantial proportion of my holding over 2 days, with some big sales being unmatched by purchases. This happened only while the account was set to withdraw repayments and interest. * Most/all loans will have repayments in a month, but on different days, so their balances will tend to move in tandem. Having sales every time one of the loans makes a repayment seems unnecessary, especially if reinvestment is not immediate. There should be a threshold test to prevent such sales until a cumulative limit (which could probably be as low as 21%) is breached. For all I know there might be one already. I've not dug that deep.
|
|
bugs4me
Member of DD Central
Posts: 1,845
Likes: 1,478
|
Post by bugs4me on Feb 14, 2015 9:23:03 GMT
As of today, the GEIA would be unable to achieve it's objectives regarding the 7% projected without a fair amount of luck being thrown into the pot. A £100 investment would only be able to acquire £40 worth of investments - there's only a couple of W/T available so the 20% rule would presumably kick in and the £60 balance would be left 'idle'.
With nothing on the horizon in the upcoming loans tab, you are therefore dependant upon others wishing to offload their green holdings on the AM. I suspect the influx of cash deposited into the GEIA over the past few weeks has effectively dried up the available investment pot. Good for those wishing to move out of their WT holdings but not good for those wishing to invest in them via the GEIA.
I've virtually sold all of my holding as it was never going to be able to use the total investment I'd made. The best I achieved was in the region of 69% and at least one investment had breached that 20% rule. Presumably that would have been adjusted down so I would have finished up with 60% invested and 40% idle.
If AC were able to source a continual flow of WT then the GEIA would no doubt be worthwhile but as things stand it doesn't look that great IMO.
|
|
|
|
Post by stuartassetzcapital on Feb 14, 2015 17:32:53 GMT
We have got through £7m of WT loans recently and the next batch are needed to assist new entrants - this only happened a few days ago and will be solved shortly by new loans. andrewholgate will advise asap as there is a pipeline imminent. chris is also looking at technical solutions. And you may have seen, underwritten loans and now drawn down loans are here in volume and that looks like a permanent fix that we have now put in place given the future pipeline also. February will be a large month, probably largest ever and March will be considerably larger again. |
|
bugs4me
Member of DD Central
Posts: 1,845
Likes: 1,478
|
Post by bugs4me on Feb 14, 2015 19:27:39 GMT
We have got through £7m of WT loans recently and the next batch are needed to assist new entrants - this only happened a few days ago and will be solved shortly by new loans. andrewholgate will advise asap as there is a pipeline imminent. chris is also looking at technical solutions. And you may have seen, underwritten loans and now drawn down loans are here in volume and that looks like a permanent fix that we have now put in place given the future pipeline also. February will be a large month, probably largest ever and March will be considerably larger again. Thanks for the feedback stuartassetzcapital - so it's all to do with timing or is it luck as to when you jump in or indeed top-up. Nonetheless, with the 20% ruling in place, in the absence of any other loans available, then effectively you are watching idle money which could be put to better use. Personally I would have thought the easiest way round it during a WT famine period would be to ignore the 20% rule and adjust back to that percentage or as near as possible as new loans become available. This would achieve a twofold benefit whereby all of my funds would have been earning the projected 7% rather than seeing in excess of 30% earning zero. Also it would mop up those wishing to offload WT units on the AM but being silently restricted by that 20% rule. Anyway, all academic to me now as I've virtually liquidated all of my GEIA holdings apart from £39.99. So whilst the 20% rule may have seemed a good idea when there were plenty of WT units available I feel it has backfired - well in my case - when there is a WT AM famine. Maybe there will be plenty of WT units availble this month or next but sooner or later there will be another shortage.
|
|
bigfoot12
Member of DD Central
Posts: 1,817
Likes: 816
|
Post by bigfoot12 on Feb 14, 2015 23:31:21 GMT
Personally I would have thought the easiest way round it during a WT famine period would be to ignore the 20% rule and adjust back to that percentage or as near as I don't know but perhaps the 20% rule is there to protect the provision fund as much as the individuals?
|
|
ramblin rose
Member of DD Central
“Some people grumble that roses have thorns; I am grateful that thorns have roses.” — Alphonse Karr
Posts: 1,370
Likes: 857
|
Post by ramblin rose on Feb 15, 2015 10:33:02 GMT
Personally I would have thought the easiest way round it during a WT famine period would be to ignore the 20% rule and adjust back to that percentage or as near as I don't know but perhaps the 20% rule is there to protect the provision fund as much as the individuals? Very likely. But it's been causing problems with the algorithms from day one, and it's silly. The 20% rule shouldn't be a rule at all, but an aim to reach when possible. Surely the process should invest as much as possible, in as wide a spread of loans as are available, aiming to redistribute as necessary when more loans are available to spread an account across. In these early days, sure it might put the provision fund at a bit of risk, when GEIA accounts are spread across too few loans, but if AC are as confident as they claim about a suitable flow of loans for this account, then they ought to be confident enough to underwrite that for the short while it will take to broaden out. As they don't let any investor easily work out what holdings they have in the GEIA, nobody should be needing to worry about 20% limits, and if the daft rule was dropped so that the GEIA accounts got fully invested, then everybody probably would stop worrying about it.
|
|
bugs4me
Member of DD Central
Posts: 1,845
Likes: 1,478
|
Post by bugs4me on Feb 15, 2015 10:33:35 GMT
Personally I would have thought the easiest way round it during a WT famine period would be to ignore the 20% rule and adjust back to that percentage or as near as I don't know but perhaps the 20% rule is there to protect the provision fund as much as the individuals? I'm sure the 20% is there to protect something. But assuming no one was trying to sell off any units this morning, then the best you could achieve would be 40% of any funds allocated to your GEIA live. The balance would be idle. There is another WT on the horizon for a small amount so you may get a 'slice of the action' with that one but I would imagine there will be other GEIA accounts also trying to balance the books elsewhere. So unless you can get 100% invested then the projected 7% is not going to be obtained ATM. It's a shame as I saw the GEIA as a passive investment but in my case assuming there was no further movement - which I doubt there would have been as there is simply nothing else around unless an investor sells off parts of their GEIA holding - then the best I could have achieved in my case would have been c4.83%. Must confess I didn't understand all the background activity with the algorithm selling and then buying pennies of the same loan all within a few minutes. Sure though that AC will get there in the end - just not today though IMO.
|
|
|
|
Post by stuartassetzcapital on Feb 15, 2015 16:04:14 GMT
Personally I would have thought the easiest way round it during a WT famine period would be to ignore the 20% rule and adjust back to that percentage or as near as I don't know but perhaps the 20% rule is there to protect the provision fund as much as the individuals? That is correct, well spotted, and the PF in turn protects the individuals.
|
|
|
|
Post by stuartassetzcapital on Feb 15, 2015 16:06:05 GMT
I don't know but perhaps the 20% rule is there to protect the provision fund as much as the individuals? Very likely. But it's been causing problems with the algorithms from day one, and it's silly. The 20% rule shouldn't be a rule at all, but an aim to reach when possible. Surely the process should invest as much as possible, in as wide a spread of loans as are available, aiming to redistribute as necessary when more loans are available to spread an account across. In these early days, sure it might put the provision fund at a bit of risk, when GEIA accounts are spread across too few loans, but if AC are as confident as they claim about a suitable flow of loans for this account, then they ought to be confident enough to underwrite that for the short while it will take to broaden out. As they don't let any investor easily work out what holdings they have in the GEIA, nobody should be needing to worry about 20% limits, and if the daft rule was dropped so that the GEIA accounts got fully invested, then everybody probably would stop worrying about it. We could change this but as it is a change to the disclosed rules then we may need to issue a Series 2 of the GEIA with new rules. We are debating this at present.
|
|
tonyr
Member of DD Central
Posts: 477
Likes: 258
|
Post by tonyr on Feb 15, 2015 16:39:05 GMT
Very likely. But it's been causing problems with the algorithms from day one, and it's silly. The 20% rule shouldn't be a rule at all, but an aim to reach when possible. Surely the process should invest as much as possible, in as wide a spread of loans as are available, aiming to redistribute as necessary when more loans are available to spread an account across. In these early days, sure it might put the provision fund at a bit of risk, when GEIA accounts are spread across too few loans, but if AC are as confident as they claim about a suitable flow of loans for this account, then they ought to be confident enough to underwrite that for the short while it will take to broaden out. As they don't let any investor easily work out what holdings they have in the GEIA, nobody should be needing to worry about 20% limits, and if the daft rule was dropped so that the GEIA accounts got fully invested, then everybody probably would stop worrying about it. We could change this but as it is a change to the disclosed rules then we may need to issue a Series 2 of the GEIA with new rules. We are debating this at present. If you publish a decent API then we can have competitions as to which algorithm give the highest return, has the highest Sharpe ratio or has the greatest liquidity. That would be fun. You might like to host the winners.
|
|
|
|
Post by chris on Feb 15, 2015 17:01:10 GMT
We could change this but as it is a change to the disclosed rules then we may need to issue a Series 2 of the GEIA with new rules. We are debating this at present. If you publish a decent API then we can have competitions as to which algorithm give the highest return, has the highest Sharpe ratio or has the greatest liquidity. That would be fun. You might like to host the winners. Both the public API and bespoke investment accounts will be in beta testing in the next few weeks. I expect them both to be live in April at the latest, and these are two of my highest priorities at the moment.
|
|
|
|
Post by andrewholgate on Feb 16, 2015 12:45:18 GMT
We have got through £7m of WT loans recently and the next batch are needed to assist new entrants - this only happened a few days ago and will be solved shortly by new loans. andrewholgate will advise asap as there is a pipeline imminent. chris is also looking at technical solutions. And you may have seen, underwritten loans and now drawn down loans are here in volume and that looks like a permanent fix that we have now put in place given the future pipeline also. February will be a large month, probably largest ever and March will be considerably larger again. Thanks for the feedback stuartassetzcapital - so it's all to do with timing or is it luck as to when you jump in or indeed top-up. Nonetheless, with the 20% ruling in place, in the absence of any other loans available, then effectively you are watching idle money which could be put to better use. Personally I would have thought the easiest way round it during a WT famine period would be to ignore the 20% rule and adjust back to that percentage or as near as possible as new loans become available. This would achieve a twofold benefit whereby all of my funds would have been earning the projected 7% rather than seeing in excess of 30% earning zero. Also it would mop up those wishing to offload WT units on the AM but being silently restricted by that 20% rule. Anyway, all academic to me now as I've virtually liquidated all of my GEIA holdings apart from £39.99. So whilst the 20% rule may have seemed a good idea when there were plenty of WT units available I feel it has backfired - well in my case - when there is a WT AM famine. Maybe there will be plenty of WT units availble this month or next but sooner or later there will be another shortage. Yes there are more deals to come.
|
|
|
|
Post by bracknellboy on Feb 21, 2015 18:08:32 GMT
Boy, the alg. on this GEIA really is bit totally bonkers isn't it ? (Or at least the outcomes appear to be...)
I'd deliberately stayed clear of it until the issues with the account were thrashed out. But a few days ago decided to dip my toe in as a bit of an experiment. like others i can see it having a use as a 'holding account', given that I could be comfortable with a level of 'over exposure' to loans I hold if the 'over' bit has provision fund backing.
But I could see it would be problematic to get invested when there appeared to be only 2 qualifying loans avaialble. So as an initial experiment I put in £100. I expected it to either invest £40 across those two loans, or initially fully invest to the £100 mark. It did neither. It invested £80 (2x£20 in each of the 2 available), leaving only £20 stranded. It settled at that. So I then moved the £20 out to cash account to see what happened. it stayed at the £80 invested level. So far so good as an experiment in how to manipulate it to get fully invested, albeit one that shed not a jot on what the rules were for the initial investment.
A few days later I've gone back to check. Its dropped the investment by about £2.50 or something. But below the surface, the duck has been paddling like mad, simply to go round in circles.
The day after my initial activity it decided to sell £24 of loan A. Presumably it had woken up to being 'more than 20% exposed' and had an opportunity to divest (£16 left would be 20% of £80). Except that 2 mins later it bought £12 of it back again. Now the aggregate of the £16 and the £12 makes no sense to me. £28 is neither one thing or the other compared to either the £80 or the sum of the £40 in the ohter loan and the £28. I could understand it may have been trying to sell some of loan B and when it couldn't it decided to buy some of loan A back again, maybe. But then it sold that £12 again 2 mins later. Then bought it back again, then sold it again and then.....get the picture.
Eventually there was a settled net result of the sale of £24 ...but only for 4 minutes. It then decided to change tack. It had obviously got bored with £12 chunks and may have decided to start limit testing the famous 20 decimal places. So instead it bought £4.53 back, oh and £7.47, two transactions with same time stamp. Ah, that's a total of £12 then.
It was then quite content for a while...a total of 12 minutes in fact (now I'm beginning to get scared, I've never liked the number 12), when it decided to sell 26p (still of loan A I hasten to add). And then another 2 mins later sold £11.74. Yep that'
s a total of £12 again. Which of course it bought back again 1 minute later. In one chunk.
It was than happy for a good hour and 20 mins, at which point thankfully it found some new toys to play with (AKA 'shrapnel') to help it genuinely diversify.
36 hours later or thereabouts, it decided to play with Loan A again, shedding £12 of it (in different size chunks), before going into a loop of buying back and selling £6 of it, eventually leaving a net position of £6 sold before finding some other toys to play with.
Somewhere sitting in here is some logic. I'm minded to downlaod the transactions and see if I can fathom it. I can see that maybe it needs to sell to free up cash before checking whether it can buy, and then partially reversing that when it can't get anything. And there seems to be some form of almost 'binary chop' going on.
I'd be intriqued to know what all this selling and buying does for interest accrual on the amounts being churned. Is it based on net balance in a loan at the end of each day ?
Edit: Loan B it left untouched (but this one has a significantly bigger pot of available units).
|
|
bugs4me
Member of DD Central
Posts: 1,845
Likes: 1,478
|
Post by bugs4me on Feb 21, 2015 21:03:35 GMT
Boy, the alg. on this GE IA really is bit totally bonkers isn't it ? (Or at least the outcomes appear to be...) I'd deliberately stayed clear of it until the issues with the account were thrashed out. But a few days ago decided to dip my toe in as a bit of an experiment. like others i can see it having a use as a 'holding account', given that I could be comfortable with a level of 'over exposure' to loans I hold if the 'over' bit has provision fund backing. But I could see it would be problematic to get invested when there appeared to be only 2 qualifying loans avaialble. So as an initial experiment I put in £100. I expected it to either invest £40 across those two loans, or initially fully invest to the £100 mark. It did neither. It invested £80 (2x£20 in each of the 2 available), leaving only £20 stranded. It settled at that. So I then moved the £20 out to cash account to see what happened. it stayed at the £80 invested level. So far so good as an experiment in how to manipulate it to get fully invested, albeit one that shed not a jot on what the rules were for the initial investment. A few days later I've gone back to check. Its dropped the investment by about £2.50 or something. But below the surface, the duck has been paddling like mad, simply to go round in circles. The day after my initial activity it decided to sell £24 of loan A. Presumably it had woken up to being 'more than 20% exposed' and had an opportunity to divest (£16 left would be 20% of £80). Except that 2 mins later it bought £12 of it back again. Now the aggregate of the £16 and the £12 makes no sense to me. £28 is neither one thing or the other compared to either the £80 or the sum of the £40 in the ohter loan and the £28. I could understand it may have been trying to sell some of loan B and when it couldn't it decided to buy some of loan A back again, maybe. But then it sold that £12 again 2 mins later. Then bought it back again, then sold it again and then.....get the picture. Eventually there was a settled net result of the sale of £24 ...but only for 4 minutes. It then decided to change tack. It had obviously got bored with £12 chunks and may have decided to start limit testing the famous 20 decimal places. So instead it bought £4.53 back, oh and £7.47, two transactions with same time stamp. Ah, that's a total of £12 then. It was then quite content for a while...a total of 12 minutes in fact (now I'm beginning to get scared, I've never liked the number 12), when it decided to sell 26p (still of loan A I hasten to add). And then another 2 mins later sold £11.74. Yep that' s a total of £12 again. Which of course it bought back again 1 minute later. In one chunk. It was than happy for a good hour and 20 mins, at which point thankfully it found some new toys to play with (AKA 'shrapnel') to help it genuinely diversify. 36 hours later or thereabouts, it decided to play with Loan A again, shedding £12 of it (in different size chunks), before going into a loop of buying back and selling £6 of it, eventually leaving a net position of £6 sold before finding some other toys to play with. Somewhere sitting in here is some logic. I'm minded to downlaod the transactions and see if I can fathom it. I can see that maybe it needs to sell to free up cash before checking whether it can buy, and then partially reversing that when it can't get anything. And there seems to be some form of almost 'binary chop' going on. I'd be intriqued to know what all this selling and buying does for interest accrual on the amounts being churned. Is it based on net balance in a loan at the end of each day ? Edit: Loan B it left untouched (but this one has a significantly bigger pot of available units). Welcome to the GEIA - yes it is bonkers in my experience. I never did fathom out what it was doing and after a few days cashed in. Still got about 30p left. There were only 3 WT's available but it used 69% on my funds so in effect broke it's own 20% rule. It stayed at the 69% mark but in the background there were pennies of this and pennies of that being sold then what appeared to be a few seconds later it was buying them back again but often in smaller amounts. As the 20% rule is still in place and there was only one WT on the horizon I decided that it was never going to get to 100% investment unless by chance others became available on the AM. Presumably the projected 7% only applies to invested funds so I would have missed that by a couple of percentage points if everything stayed as it was. What the GEIA has managed to achieve is to mop up all those WT's that were on the AM and probably no doubt many lenders have 100% invested as a result. But IMO the 20% rule should be relaxed at times of famine. So not for me ATM.
|
|
|
|
Post by bracknellboy on Mar 2, 2015 18:19:50 GMT
chris: back to a question I raised on my original post. With all this churning (intra day on same loans) what is the score w.r.t. recognition for interest ? Is it based on net holding at end of each day ? Want to be sure that the constant sell/buy of loan parts of the same loan multiple times a day is not impacting negatively on the interest received. [Note:I currently have only a trival £80 holding in GEIA and am not minded to put any more in until I understand its operation more].
|
|
