The Financial News reports that Goldman Sachs has become the first bank to create a hedge fund replication tool in a move that could lead to a shake-up of the $1.3 trillion (€976bn) hedge fund industry.
The article reveals that the platform will greatly undercut the notoriously high fees of the hedge fund sector. Those investing through a fund of funds can end up paying annual charges of 4% to 7%, with up to 50% of their returns eaten up by fees. Goldman will charge a flat 1%.
Goldman has spent two years developing the algorithm that underpins its platform. The performance characteristics of thousands of hedge funds will be fed into the system monthly and Art is designed to decompose these data and calculate the aggregate position of the hedge fund universe, reports the article
Goldman is joining Dow Jones and Merrill Lynch in the race for market share in the nascent business of hedge fund "clones" that are aiming to replicate hedge fund returns but with much lower fees. Last week, Professor Harry Kat of the Cass Business School at the City of London University launched an online venture called FundCreator. The synthetic fund uses 78 futures contracts to model different risk-return profiles and envisages charging just 0.36% per annum in fees. It does, however, require about $20 million to fund the model, though that hasn't stopped around 10 investors from testing the system.
Kat, Goldman and copulas
Kat, Goldman and copulas
Please correct me if I'm wrong, but when I read the paper, I think I see that K&P are replicating a FOHF return stream after both sets of fees have been deducted. Rather than giving those fees back to the investor, they are flushing them down the toilet.
What they could have attempted, but apparently didn't, is this:
What they could have attempted, but apparently didn't, is this:
- Put the FOHF returns, after fees, into a spreadsheet
- Calculate what the pre-fees returns would have been, to produce the post-fee returns in step 1.
- Pretend the pre-fee returns in step 2, are the returns of a fictional FOHF, "Generous Management Fund" (GMF)
- Use the copula approach to replicate the returns of GMF
- Attachments
-
- Copula.pdf
- (1.53 MiB) Downloaded 676 times
-
AFJ Garner
- Roundtable Knight
- Posts: 2071
- Joined: Fri Apr 25, 2003 3:33 pm
- Location: London
- Contact:
I was passed a copy of an earlier and briefer Kat paper last year by Morgan Stanley on the same topic.
Yes, I get the idea that you can replicate FoF returns using S&P 500 futures and (10 Yr?) Treasury futures.
But if you search amongst Kat's Klingon Formulae and Copula(tion?) approaches, you are most unlikely to be left any the wiser as to how his strategy works on a day to day basis.
Put it this way, if you wade your way through all 52 pages of the attached paper, you may very well NOT find yourself in a position to start coding his strategy into TB.
Well who said life should be easy?
Yes, I get the idea that you can replicate FoF returns using S&P 500 futures and (10 Yr?) Treasury futures.
But if you search amongst Kat's Klingon Formulae and Copula(tion?) approaches, you are most unlikely to be left any the wiser as to how his strategy works on a day to day basis.
Put it this way, if you wade your way through all 52 pages of the attached paper, you may very well NOT find yourself in a position to start coding his strategy into TB.
Well who said life should be easy?
Excerpt from the Kat and Palaro paper:
I guess that strategies that exhibit positive skewness (e.g. Managed Futures) are safe from the replicants!"We also see that the replication strategy is unable to replicate the three large losses that LCH reported in October 1987 (-22.52%), August 1998 (-11.45%) and April 2000 (-10.83%). Since these are clearly outliers, it is not surprising that the replication procedure was unable to capture them out-of-sample. Given the size of these losses, it is unlikely investors will consider this a real shortcoming though."
Sluggo,
I agree with you, I guess that selling point for the synthetic fund is (quote from the pdf) the ability to generate very similar returns in a much more mechanical way and with a lot less effort. If it is, we may be able to do without expensive hedge fund managers and all the hassle, including the due diligence, the lack of liquidity, the lack of transparency, the lack of capacity, and the fear for style drift, which comes with investing in hedge funds.
It looks like mutual funds vs ETFs
or Kat is trying to become a fat cat???
I was trying to see if I could code the ideas into a trading system, but even AFJ Garner couldn't so I give up (...and I will keep on copulating in the old fashioned way).
best regards, as ever
I agree with you, I guess that selling point for the synthetic fund is (quote from the pdf) the ability to generate very similar returns in a much more mechanical way and with a lot less effort. If it is, we may be able to do without expensive hedge fund managers and all the hassle, including the due diligence, the lack of liquidity, the lack of transparency, the lack of capacity, and the fear for style drift, which comes with investing in hedge funds.
It looks like mutual funds vs ETFs
or Kat is trying to become a fat cat???
I was trying to see if I could code the ideas into a trading system, but even AFJ Garner couldn't so I give up (...and I will keep on copulating in the old fashioned way).
best regards, as ever
Maybe at an academic conference somebody will ask, "Professor Kat, could hedge funds or funds-of-hedge-funds lower their fees and thereby improve their net performance to a level such that your technology cannot replicate the returns?"
If he answers "yes" then he has identified an Achilles Heel of the approach, and (some) funds and FOHF's will lower their fees.
If he answers "no" then the natural follow up question is "Suppose they lowered their fees to ZERO. Can you replicate their performance if their fees are Zero." If so then why hasn't he presented this in the first place, as it would be of tremendously greater economic value to investors; has Goldman suppressed these results?
If he answers "yes" then he has identified an Achilles Heel of the approach, and (some) funds and FOHF's will lower their fees.
If he answers "no" then the natural follow up question is "Suppose they lowered their fees to ZERO. Can you replicate their performance if their fees are Zero." If so then why hasn't he presented this in the first place, as it would be of tremendously greater economic value to investors; has Goldman suppressed these results?
-
AFJ Garner
- Roundtable Knight
- Posts: 2071
- Joined: Fri Apr 25, 2003 3:33 pm
- Location: London
- Contact:
Passive equity funds which track stock indices, be they exchange traded or of the mutual fund variety, have many advantages over actively managed funds. Few would seek to argue the point, although many would point out the superior performance of a small number of the latter.
Two of the biggest advantages of index trackers are simplicity and transparency: you can easily see exactly how the returns are derived and a well run fund will have only the smallest of tracking errors relative to its given benchmark.
Does this fact put the managers of such funds out of business? No, Vanguard still seems to manage well, despite the fact that theoretically anybody with time, a little intelligence and lots of money can track an index of their choice.
So, instead of asking Kat the questions suggested by Sluggo, I would ask him to be more explicit about his strategy so that potential investors, given enough energy and the right tools, can see for themselves, from their own testing, what returns can be replicated and how.
After all, one of Kat's stated advantages for his method is "transparency".
So, he could make his point a lot more clearly with a worked example or two. In order to duplicate the returns of this or that hedge fund, what long positions or short positions in the S&P, T Bills and the reserve asset were made each day? And exactly how were such daily trading decisions arrived at? Given enough time, no doubt you could approximate his system for yourself but if his aim IS transparency, then why not make things a little simpler?
After all, as with index trackers a) if his method is so valuable and simple it won't be long before rivals cotton on and design their own products; but b) few will actually want to do it for themselves, so he should still have plenty of potential investors left.
Unless the methodology is made crystal clear, the position of an investor in the Goldman's ART fund or Kat's product will be little different to that of an investor in a bog standard hedge fund. It would be nice to understand just how your money is being invested (or pissed up the wall).
Given the current state of disclosure, an investor may just as well buy an undisclosed futures trading system and follow its little green and red lights.
Kat and Goldman's are undoubtedly pointing in the right direction. But unless their methods are as transparent and clear as that of the index tracking equity funds they aim to emulate, I can not see that they have any very clear advantage over the more traditional hedge fund manager.
You can believe (or not) the copulatory mumbo jumbo put out by system sellers and Kat and Goldman as you choose.
But given the current state of disclosure, I wonder whether you will have any more idea as to what Goldman or Kat will be doing with your money than you would have done with Amaranth or Long Term Capital Management?
Two of the biggest advantages of index trackers are simplicity and transparency: you can easily see exactly how the returns are derived and a well run fund will have only the smallest of tracking errors relative to its given benchmark.
Does this fact put the managers of such funds out of business? No, Vanguard still seems to manage well, despite the fact that theoretically anybody with time, a little intelligence and lots of money can track an index of their choice.
So, instead of asking Kat the questions suggested by Sluggo, I would ask him to be more explicit about his strategy so that potential investors, given enough energy and the right tools, can see for themselves, from their own testing, what returns can be replicated and how.
After all, one of Kat's stated advantages for his method is "transparency".
So, he could make his point a lot more clearly with a worked example or two. In order to duplicate the returns of this or that hedge fund, what long positions or short positions in the S&P, T Bills and the reserve asset were made each day? And exactly how were such daily trading decisions arrived at? Given enough time, no doubt you could approximate his system for yourself but if his aim IS transparency, then why not make things a little simpler?
After all, as with index trackers a) if his method is so valuable and simple it won't be long before rivals cotton on and design their own products; but b) few will actually want to do it for themselves, so he should still have plenty of potential investors left.
Unless the methodology is made crystal clear, the position of an investor in the Goldman's ART fund or Kat's product will be little different to that of an investor in a bog standard hedge fund. It would be nice to understand just how your money is being invested (or pissed up the wall).
Given the current state of disclosure, an investor may just as well buy an undisclosed futures trading system and follow its little green and red lights.
Kat and Goldman's are undoubtedly pointing in the right direction. But unless their methods are as transparent and clear as that of the index tracking equity funds they aim to emulate, I can not see that they have any very clear advantage over the more traditional hedge fund manager.
You can believe (or not) the copulatory mumbo jumbo put out by system sellers and Kat and Goldman as you choose.
But given the current state of disclosure, I wonder whether you will have any more idea as to what Goldman or Kat will be doing with your money than you would have done with Amaranth or Long Term Capital Management?
I agree, it would have been too nice to have a Trading Blox-style back-testing report.AFJ Garner wrote:So, instead of asking Kat the questions suggested by Sluggo, I would ask him to be more explicit about his strategy so that potential investors, given enough energy and the right tools, can see for themselves, from their own testing, what returns can be replicated and how.
and from BusinessWeek:
Hedge Funds: Attack of the Clones
Indexing may not appeal to current hedge fund investors, though, some analysts observe. "The typical hedge fund investor may not wish to effectively trade in their Ferrari for a potentially more reliable—and arguably boring—Buick," says Jeff Keil, principal at fund consultancy Keil Fiduciary Strategies.
Kat has a new website: http://www.fundcreator.com/