Ms Naivete,
you wrote:
what is the best protection against this kind of crashes/scams like LTCM, etc?
Glad you asked.
Short answer: There is none.
Long answer:
There is a nice book about the LTCM story:
Roger Lowenstein: “When Genius Failed: The Rise and Fall of Long-Term Capital Management”
(http://www.amazon.com/When-Genius-Failed-Long-Term-Management/dp/0375758259/ref=sr_1_1?ie=UTF8&s=books&qid=1295181480&sr=8-1)
The books is a nice read and shows that there is much more to the story than an excellent trading strategy which failed spectacularly after a few years.
For some of the root causes I’d sugest an other favourite book of mine:
Nassim Nicholas Taleb: “The Black Swan”
(http://www.amazon.com/Black-Swan-Improbable-Robustness-Fragility/dp/081297381X/ref=ntt_at_ep_dpi_1)
Its a very enjoyable read. Taleb tries to explain his point with many every-day examples, a must-read for every trader. I doubt there are many people who understand him however. We humans are not very good when it comes to statistics and rare events. Things are usually counter-intuitive there.
In physics we talk about complex systems (http://en.wikipedia.org/wiki/Complex_systems). These are systems with non-linearities and feedback-loops etc. Some examples include Earthquakes, Avalanches and financial market crashes.
If you go through the theorie you’ll discover, that there is no prediction of and no measure against such events in priciple.
Thanks Koch!
Anybody can read the books, especially these popular science ones, you have mentioned.
But are you familiar with some recent work done on the issue at the Santa Fe Institute?
If you are not familiar with it, let me know and I will post the reference here for you.
Now, why the LCTM was not a scam, as the folks would call it here, on C2, upon its crash?
You wrote:
are you familiar with some recent work done on the issue at the Santa Fe Institute
No, I am not, please enlight me…
why the LCTM was not a scam
Nobody knows exactly what was going on, but from the book mentioned above I learned this:
They had a working strategy, backed by solid quant research, simulations, conducted by very clever people.
But then, suddenly, the real world started to behave differently than their models would have predicted. Price movements went in the opposite directions, all of them, at the same time.
An event that may happen once in 100 years. They simply did not think about such an event before it occured.
That is exactly what N.N. Taleb calls “The black swan”. Nobody thought there was a black swan before the first was discovered.
The same story (somewhat scaled down) happpened several times to every mature trading system developer. Your freshly developed system works very nice for all situations in the past - and fails not very long into the future because the markets create a situation for you which goes completely against all your assumptions.
That is why people are right who want to see that a system survives for a longer time, can handle different market situations, especially situations which are new and unexpected.
Remember October 2008? They called it a century crash.
Try to find stock systmes on C2 which survived October 2008 - there aren’t many.
This is my answer to the first question in this thread of postings:
Q: "what makes a system exceptional?"
A: If it can handle situations it hasn’t seen before.
Ms Naivete, why don’t you just post the reference you are referring to instead of trying to make Rene look uneducated because he hasn’t seen the latest from the Santa Fe Institute? I haven’t either and, although I’m sure it is very interesting (I mean that sincerely - the Santa Fe Institute was where I wish I had done my PhD back in the 90’s when that was the place to be) I doubt it would have much effect on my day-to-day trading.
One of the problems the LTCM wizards had was exactly that they relied too much on the “latest” complex theory at the expense of wisdom and common sense. Yes anyone can read Taleb but the most interesting point is that people still make the same errors even though they can pick up a bestseller telling them about it. And Taleb’s ideas aren’t exactly new either; but he does write well and his books are very good indeed.
There is nothing wrong with the latest complex theory. It is the way science works - theories are developed, flaws are usually found later, the theories are either modified or discarded. It is a very fluid process which is something that the majority of non-scientists do not fully appreciate. The problem arises when complicated latest theories are taken on faith and applied blindly to the real world, e.g. trading.
As to whether LTCM was a scam, I don’t believe so. They were just misguided, as I said above. I believe most of the C2 boom-bust systems are also just misguided. But when a vendor does the same again and again, often under a different alias, then that is a scam.
Rene has made several very good points indeed. From following his posts over the months anyone with intelligence and common sense can see that he belongs to the section of C2 members who have their heads screwed on. Better to to take serious note of the things he says instead of treating him as another scammer.
Thanks Dean for the nice words.
I believe many of the LTCM principles launched JWM Partners in 1999. They followed many of LTCM’s strategies but with less leverage. JWM Hedge Fund was shut down in July 2009 after a 40+% loss.
Rafael,
In hindsight, I was able to easily find systems that met your criteria for a âGood systemâ.
I simply looked at the âlosers this weekâ and âlosers this monthâ section on C2’s dashboard. Many of these systems met your criteria in their 30 days of existence.
Certainly, I will.
Let me start by saying that I am not here, on C2, to promote somebodyâs system or disparage other systems. I am here to study trading systems and build portfolios of them, applying publicly available results of current financial research programs.
Initially, I had thought that building of profitable portfolios of systems was the purpose of C2 existence.
Instead, I notice rather dysfunctional behavior on C2: predictions of system crashes, harsh accusations of scams, etc.
Next, The Santa Fe Institute is a private, not-for-profit, independent research and education center founded in 1984. Renowned scientists and researchers come to Santa Fe Institute from universities, government agencies, research institutes, and private industry to collaborate in attempts to uncover the mechanisms that underlie the deep simplicity present in our complex world.
I find that a working paper âOPTIMAL LEVERAGE FROM NON-ERGODICITYâ by Dr. Ole Peters, is very helpful in my work. Ole comes from Imperial College in London. He is a very talented researcher, known worldwide.
You can view it at
http://www.santafe.edu/media/workingpapers/09-02-004.pdf
The popular books you had listed are fun and easy reading.
Great for entertainment purposes!
However, when it comes to serious work on systems evaluation, these books offer little help.
Ole’s paper is not as easy to read but is fun too, and it is right on the money.
It looks at portfolios from an angle that is very useful to our system evaluation purposes.
It concludes, for instance, that it only makes sense to compare portfolios on a time scale much longer than five years.
So, practically, all C2 systems are still too young to be compared to each other.
The paper also concludes that high returns with low return volatility (preferably, low draw downs) are desirable.
Therefore, if one choses to get involved with systems that are younger than five years, it pays to go for a system with a high return accompanied by an acceptable draw down.
There are other conclusions presented in the paper as well.
I can list them for anyone who is interested in the paper but do not care to dig into the math used in it.
By the way, Professor Taleb, you have mentioned, calls himself a mathematical trader.
Hopefully, the C2 investment community will find some functional ways of trading systems evaluation in the near future.
I think this is a very useful thread – and agree that it would be great if C2 and its members helped to devise good, functional, ways of evaluating trading systems.
There’s a fine balance between results: short-term results/young systems on C2 – and longer-term system evaluation.
[LINKSYSTEM_51201525]
Rafael,
Can you please tell me, and I don’t want to be sarcastic but I mean it in all sincerity, how many subscribers at 100% scaling do you think your system “Stocks Hunter” can support before slippage would kill it - given that it trades many penny stocks and other very thinly traded stocks?
In addition, whenever someone considers subscribing to a penny stock system one has to ask himself does the vendor have your best interest in mind or is he just putting the system up so he can front run it.
I am not implying that this is you intention but these are the questions potential subscribers have to ask themselves.
The challenge is that each person’s demand for performance and tolerance for risk (greed and fear threshold) is different. One might argue that many people have a completely unrealistic performance vs. risk demands (they want 200% return/year with less than 10% drawdowns).
The real world just doesn’t work that way.
Spectacular returns = spectacular risk
There’s no way around that. The risk may be hidden, in the short term, by a string of good luck, but the law of large numbers will catch up with everyone, eventually.
I’ve read a number of hedge-fund performance reports, which indicate that a CALMAR Ratio (three-year CAGR/max. drawdown) of 3 or more is “exceptional” results. If you have a system, which has a 100% CAGR, that means you can get by with a max drawdown of 33% and still be considered “exceptional”.
How many people, who chase after those 200% CAGR systems would be willing to tolerate a 67% drawdown before declaring the system as “broken” or “crashed”?
On the other hand, in order to get the max drawdown down to the desired 10% (for an “exceptional” system), who would be willing to accept a 30% CAGR?
Systems in these ranking of "losers" can not fit my criteria of good systems (Age > 30, Ann return > 100, Max DrowDown < 30). Of course, this is a simple criteria, a first filter. More details must be considered.
Anyway, what I mean is that there may be young systems on Collective2 that are good and they will be good (and probably more that some of the older ones). And many of these systems are new on Collective2 but they have a track record or backtesting results out of here.
In my case, I met a month ago Collective2, but my system
[LINKSYSTEM_55660543]
has been publicly audited since more than 6 months on my website, and of course, is the result of years of research and work.
I understand that there are users who do not trust the system "so young" like mine, but with these modest numbers I can not say it is not a good system.
Regards.
Rafael,
When you build systems do you look at more than the numbers? There have been many systems that have met your criteria. Suppose that a system that exceeds your criteria averages down like enchante,would you consider it?
What about a system that looks for the strongest trends within a group and swing trades that …but only trading with the trend.
Given a black swan can hit either. Which would you prefer?
Reinar.
Rafael,
Sorry if I was not clear.
1) Take a look at the systems that are now “losers of the week” and “Losers of the month”.
2) Now look at the first 30 days of those sytems.
3) Notice that there are many of those system that would have met your criteria after the first 30 days of their existance.
4) Finally, compute how much profit you would have made if you invested in those sytems after the first 30 days in existance.
I’m not saying your system will be one that will crash, just that there are many systems that have good statistics in the first 30 days. Very few (if any) would meet your criteria after 1 year
Karl,
Firstly, I am not sure how many subscribers my system [LINKSYSTEM_55660543] could support, but as soon I see that there is significant slippage I will not admit more subscriptions. I am the first interested in maintaining the quality of my system.
Secondly, do not confuse low-priced stocks with illiquid stocks. All the stocks my system trades are low-priced stocks but liquidity ones. My system its been running since 6 months ago (you can check it in my website) without any subscribers who "help me", as you suggest, to gain that 15%-20% I am getting every month. If that were my intention, I will set a ridiculous subscription price ($5-$10/month) to my system, instead the current $199/month, do not you think?
Regards.
For a combination of longevity and performance I believe Interlink trading is one the best programs on C2.
Bradley,
Thanks for the input. It is true that there are very few exceptional systems on C2 that have longevity too.
Chris- Sure, you have to wonder what people are thinking when they’ll throw money at a program with 3-4 months of history. The fear-greed conundrum can be hazardous to your financial health, and your physical health for that matter. Unfortunately, some people have to learn the hard way!
Yes. And unfortunately I believe a large portion of the "investors" who come to C2 have little experience and/or knowledge. In addition to not understanding the need to see at least 1 to 2 years of system performance (in most cases), it seems many of them falsely believe that high win percentage or high profit factor is the road to riches.
As with both of our systems, less than 45% of the trades are winners yet both systems have a decent CAGR and MAR (with longevity too).
It would be great if Matthew could add an educational piece (prominently displayed) to his website that could dispel the myth about high win percentage and high profit factor, and also point out that risk (drawdown) per trade is an important aspect of a system.
I am astounded by a recent My Analyst write-up about a system on C2 praising that it had no losing trades (but the writer did not realize the hidden danger signs of high/extreme trade drawdowns–an "accident" waiting to happen).
