Do any of the books go into detail explaining how and why successful strategies worked? Historically that is. I wouldn't expect any strategies outlined in a book to be profitable today, but I would be interested in how a strategy was formalised.
I prefer to pick that sort of thing up from mathematical texts or a textbook.
The principles are pretty simple from a probabilistic perspective: it's just a matter of making inferences from the available information and choosing an optimal course of action from that. Information theory provides a nice framework for this basis, imo.
You can find this introduction (including topics like Kelly's criterion etc) in Cover's Elements of Information Theory, a good introduction to information theory in general if you like.
It's so important for physics, materials science and chemistry, but feels so esoteric. It's wonderful that there's someone like Simons with both money and the ability to understand the problem :)
edit: He also possibly implies that the best way to help maths and theoretical physics is to leave the field, attempt to get rich in finance, and pump money back in.
It's sad that this is a good way to fund fundamental research. Why is it that finance make more than what seems to be more important and useful like math or science!?
Being sad about it is not so useful. You can actually help! Either becoming rich and helping basic science (of course) or voting well and influencing government decisions.
Simons is clearly very talented at what he does, his firm (how it is run, etc) seems quite amazing. And he's "doing" more science this way. It makes an awful lot of sense.
but he's not "doing" science anymore, and if i were in love with doing science, i really wouldn't want to go into finance, do lots of stuff i don't personally feel passionate about, get really rich, then pay other researchers to do science.
About Simmons specifically, he's actually published some research with a ton of applications in recent physics developments, which I think came out of his Chern-Simmons form. He had a good run in science before he switched. And he's been doing good mathematics lately it seems.
What I am talking about is mostly compassion with both fellow scientists and humanity -- thinking basic science is really important for society as a whole and enjoyable, and it should be a good career path because of that. If you consider only feeling passionate about, you could say "It's really sad people who are really passionate about Frisbee aren't paid to do so." (nothing against Frisbee of course, but you sure could do it only as a hobby! ).
Besides, Simmons has been doing what he likes and what he feels important, I think it's pretty cool. It sure doesn't hurt he's made an awful lot of money too.
This is an intellectually and morally bankrupt sentiment that is all the more worth calling out because it is both impractical and nonsensical.
1. This sentiment is nonsensical. Scientific funding often funds more than just salaries, and when it does fund salaries, those salaries are almost always modest. Furthermore, many of the salaries it funds are for non-scientists: technicians, software engineers, etc. The tired trope that scientific funding is primarily funnelled to wide-eyed loners who produce nothing but "useless" abstract ideas is demonstrably inaccurate. And to the extent that this caricature is accurate, Simons himself is a counter-point to the argument that this is a bad model.
2. This sentiment is impractical. Basic science is important to society as a whole and there's almost always no way to become ultra-rich doing basic science(++). All of society benefits when a good lot of our best and brightest go into science (rather than e.g. consulting or finance). And the best way to ensure that great minds go into science is to ensure science remains funded so that it is possible to work on truly important problems.
3. This sentiment is intellectually bankrupt. A great mind following its dreams without prioritizing financial reward is responsible for most of the major scientific developments that make it possible for you to bash their would-be descendents from behind a keyboard.
4. This sentiment is morally bankrupt. Punishing passionate people who give up highly lucrative careers to do something that is good for humanity is nothing short of vindictive -- I hate my work day so everyone else has to as well!
(edit: As an aside, you really think CEOs don't like the rush from having lots of power and making important decisions? They may work long hours, but I guarantee most of them fucking love their jobs. I also bet there's a pretty strong correlation between top-of-clas software engineers who command high salaries for their expertise, and software engineers who love their job.)
(edit2: Furthermore, science has lots of drudgery and hard, frustrating work to it; it's not all sitting in an office and drinking coffee. The idea that scientists unequivocally work on fun problems all day and never bash their heads against the wall to solve problems that they're more extrinsically than intrinsically motivated to work on is also pretty wrong.)
(++) Simons explicitly answers the question in his interview: No. Nothing we do at Renaissance -- no matter how impressive from a finance perspective -- is useful to science. It's just useful for making a handful of people rich.
I am from Stony Brook, he has done a lot for the community including a nice nature preserve but I can not read his name and not be sad. Despite all this money, his life has been very tragic, two of his three children died young, one drowning and one in a car crash.
Not only that, they once funded the collider experiments at a physics laboratory where I worked when NSF had cut funding drastically, we're talking millions of dollars here.
Also engaged in one of the most ridiculous tax dodges of all time. Uses a "derivative" with his brokers that allows a bunch of short term trades to be taxed at long term capital gains rates.
Nothing ridiculous about any legal interpretation of tax laws that allows one to minimize one's contribution. It's common sense. Even more so given the outrageous way in which government wastes the revenue it gets.
"No man in the country is under the smallest obligation, moral or other, so to arrange his legal relations to his business or property as to enable the Inland Revenue to put the largest possible shovel in his stores. The Inland Revenue is not slow, and quite rightly, to take every advantage which is open to it under the Taxing Statutes for the purposes of depleting the taxpayer's pocket. And the taxpayer is in like manner entitled to be astute to prevent, so far as he honestly can, the depletion of his means by the Inland Revenue"
Lord Clyde gave this famous quote (in taxation circles) in the case of Ayrshire Pullman Motor Services v Inland Revenue [1929] 14 Tax Case 754, at 763,764:
His Medallion Fund seems to have done really well. From another article:
Simon's flagship fund, Medallion, requires aminimum investment of several million dollars and charges a 5% management fee and a jaw-dropping 44% performance fee. The fund is closed to new investment and has returned an astounding annual average net of 38% (remember, that's after the high fees). Since its 1988 launch, the fund has lost money in only one year, 1989, which saw a drawdown of 4%.
But I have to ask: why limit the fund to multi-millionaire investors? Why not open it to the other investors too and let them benefit for a change? This is just the "rich-getting-richer". If anybody needs a 38% average annual return, it's most certainly not a multimillionaire; it's more likely a middle-class person who could use the gains.
I apologize for sounding negative here, but this inequity is something that bothers me.
a.) It's much easier to maintain high returns with lower amounts of invested capital. Warren Buffett has talked about this at length; when you're investing $5M, there are a large number of undervalued opportunities where you can gain outsized returns, but when you're investing $500B, you are the market, and it's extremely unlikely that you'll get above-average returns.
b.) Private capital with a personal relationship to the fund manager is usually more patient. Many public funds have a problem where all the individual investors head for the exits at the slightest sign of trouble, but the best investment opportunities are usually available when everybody else is panicking.
Basically, if you want to do well in investing, you really need to do it yourself, and you need to have an information advantage on everyone else in the market. It's a field that rewards being smarter than everyone else, not being more cooperative.
If it was really that easy to identify undervalued companies, you'd think more people would be doing it. I don't think many managers, or individuals, beat the market by a lot, if at all.
I agree, but people who entrust their money to others to manage beat the market by even less. Remember that on average, you should expect to earn an average return minus fees; the one part of that equation that is given are the fees.
first off, there are strict SEC rules on what constitutes a qualified investor. But second, nobody in the hedge fund world wants to hand-hold a bunch of limited partners -- the fewer the better. And if you've got a track record like Simon (and Steve Cohen and Ray Dalio and Paul Tudor Jones and DE Shaw and a few others) then you're likely way oversubscribed whenever you want to open a new fund and as a result can be very picky about whose money to take. Hence the 44% carry.
The Medallion fund is a bizarre outlier. For the most part, high expected return => high risk, which is not what you want for a middle-class wealth-generating engine. It would be much better to tax the hedge-fund winners (at the end-point, with a consumption tax), and redistribute that.
RT's limits are much higher though, and that probably is done with the goal of limiting the number of individual investors. Dealing with investors is a distraction for the fund managers. But the main issue is proprietary information. The more investors you have the higher the risk of information leakage to your competitors.
He states another reason in the interview — if the fund gets too large their models will stop working because their decisions will have too much influence on the market.
I also wonder why those limits exist more generally, not just for this fund.
Maybe those limits affect the timelines on which money is invested and in which it is withdrawn, where-as actually having rules about such things would not work as well.
Was thinking about this as well. In the interview he talks about the act of making a trade moves the market itself. You can't create too much trading volume or you will have distorted the market so much that you can't trade effectively after.
I would imagine he can move and make money at a certain volume (a bank roll of 200 million) and guarantee returns because that level won't distort. But if he had hundreds of billions he couldn't guarantee returns because the moves he would have to make would be so large as to distort the market beyond their measuring.
Much like the observer effect in physics, possibly a trader's effect here.
The simple answer to your question is this: regulation.
The U.S. Securities and Exchange Commission (SEC) prevents hedge funds, private equity firms and other private investment managers from marketing their products to a wide audience.
High risk funds, like the Medallion Fund, are limited to fund raising to an exclusive group of investors that must meet the following criteria:
a) those with a net worth of at least $1 million excluding their primary residence, or
b) annual income of more than $200,000 in each of the two most recent years.
The deeper answer to why the SEC does this is: protection of the 'lowest common denominator' investing public.
By and large the investing public are not what you might call:
1) 'professional' (keeping an eye on the market as part of their job), or
2) 'sophisticated' (as in have a knowledge of the multitude of securities, how they work and how to trade them), or
3) Have a high risk profile and matching capacity for loss.
I think these points, from the SEC's point of view, make the selling of the Medallion fund to the general public inappropriate. I would imagine that from the SEC's perspective, the Medallion fund would be a 10/10 risk profile (highly speculative) and therefore the SEC would want an investor to make an investment fully understanding what they were investing in, and having enough wealth to suffer a total loss of money invested.
This leads into another point, the wording of your post implies that the 'astounding' annual average of 38% net of fees is practically a 'sure thing' when you say that since the fund's launch it took one drawdown in 1989.
However, you miss the obligatory warning that past performance is no guarantee of future results.
Speaking as a retail trader I can quite happily say that for different levels of risk I can make (and have made) the following returns (net of fees):
a) 1% per month with a high level of confidence with what I would call a very low risk trading strategy (annualised to 12% per annum),
b) 2% per month with a medium to high level of confidence with what I would call a medium risk trading strategy (annualised to 24% per annum),
c) 3% per month with a low to medium level of confidence with what I would call a medium to high risk trading strategy (annualised to 36% per annum), and
d) 4%+ per month with a low level of confidence with what I would call a high risk trading strategy.
The point I'm making here is that:
* If you know what you are doing,
* You have the capital (and capacity for loss), and
* You have the risk profile,
Then you can make decent market returns too. Technology has democratised the markets so there really isn't any reason for being negative and bothering about inequality.
You would however have to work for it and if the Medallion Fund are making 82% per annum (38% + 44%) before management fees they are working extremely hard for that.
Information on Renaissance Technologies is so incredibly sparse. It must be a fascinating workplace! It's fun to speculate on the nature of their success (specifically the Medallion fund).
3 possibilitiese spring to mind:
1. Best in class execution on winner-takes-all strategies. Eg. 10 hedge funds all try the same arbitrage trade but only the quickest one make money.
2. A killer app that no one else knows about. Like, I dunno, artificial intelligence...
3. Medallion is just a legend built up to lure investors into their public funds and collect the fees.
Joking on that last one.
What I drew from this interview was that while Simons is obviously incredibly smart, they key thing was that at Stony Brook he learned how to put together a team of other very smart people, and that more than any individual financial insight has generated his success in investing.
Simons is mentioned but he refused to be interviewed for the book. However, it covers the history that leads up to the current day.
It's interesting that Simons' fund is pure algorithmic trading. No one decides that a position is too big, for example. Humans are out of the process after they write the algorithms.
That is incredibly facinating. The way he talks about the quant models in the interview they sound simple in comparison to his prior work. My suspicion is that mathmeticians are involved in the front end model creation purely to create efficient ways of finding performant models but the models themselves are pretty mundane. The idea of purely data driven models is fascinating because it's generally rejected as too limited and inefficient. I wonder what types of anomolies they have discovered and what data sets they find useful. Also how does historical simulation work when the computer models you were competing against 10 years ago are so different than now?
As elsewhere in these comments: tax dodging is an obligation for any intelligent person.
"No man in the country is under the smallest obligation, moral or other, so to arrange his legal relations to his business or property as to enable the Inland Revenue to put the largest possible shovel in his stores. The Inland Revenue is not slow, and quite rightly, to take every advantage which is open to it under the Taxing Statutes for the purposes of depleting the taxpayer's pocket. And the taxpayer is in like manner entitled to be astute to prevent, so far as he honestly can, the depletion of his means by the Inland Revenue"
Lord Clyde gave this famous quote (in taxation circles) in the case of Ayrshire Pullman Motor Services v Inland Revenue [1929] 14 Tax Case 754, at 763,764:
Please don't be so naive. It's only because Renaissance is so rich and powerful that they are not in jail. If some piker tried the same strategies, they'd be in jail. The politicians are afraid to jail billionaires.
> The Democrat stopped short of saying that any of the activities that allowed Renaissance to lower its tax bills were illegal, and said it appears the banks stopped offering them after the IRS started to question the financial arrangements.
Wow, I love his brooklyn accent. Makes him sound so grounded. I guess Feynman has (had) a similar accent; and i was similarly enthralled hearing his voice (recording) for the first time.
Somehow I always presume these guys should speak with an English accent, like Penrose. Hehe.
Also worth reading about is his co-founder & the former chief scientist at renaissance, Nick Patterson. If there's anyone whose literally a "mad scientist" he's it (in a good way). A close friend of mine at MIT collaborates with Nick, and when we asked why he left finance to return to academia he said "well, James is into boats and he wanted a bigger one". Nonetheless, both great guys.
Anyone happen to have a transcript or alternative site to view the video from? I'm stuck behind the GFW without a VPN, and I'd love to watch this now as opposed to waiting until I get back to the States.
No, right place / right time / right choices etc still plays an overwhelming role in the wealth generation.
For every thousand geniuses that try to get really rich, there's one Simons or Buffett that pulls it off. You could run a simulation with a large number of people of equal mental capability, almost all of them would fail. Just the broader requirements alone would be enough to instantly flunk most people out of the running (the other skills required that Simons possesses beyond his intellect).
Some of this maybe a repeat to people as I often get asked for recommendations on how to get into algorithmic trading.
http://imgur.com/OdzB4aW
http://www.amazon.ca/Fortunes-Formula-Scientific-Betting-Cas... The history of hte first real quant
http://www.amazon.ca/Dark-Pools-Machine-Traders-Rigging-eboo... The history of the rise of algorithmic trading
The phyisics of wall street was mentioned by someone else, great book
http://www.amazon.ca/Quants-Whizzes-Conquered-Street-Destroy... This profiles 4 famous traders including Simons, alos a great book
http://www.amazon.ca/Heard-Street-Quantitative-Questions-Int... A must read if you want to get into quantitative finance.
As always, email if you'd like to chat.