I suggest to create a new general forum, e.g. “Trading and the Markets” where we can have discussions about general principles of trading and the behavior of markets.
Right now there isn’t a good place for such discussions: They are not suggestions to C2, they have nothing to do with autotrading, the do not pertain to a specific system, and they do not qualify as chatter.
I also suggest that if I forget to give a title to my post on this forum, and C2 forces me to go back to the previous page, that my written text will still be there and that I don’t have to write it again.
I second that wholeheartedly, on both points.
"I suggest to create a new general forum, e.g. “Trading and the Markets” where we can have discussions about general principles of trading and the behavior of markets. "
Maybe , each system designer can contribute a little bit of their experties about the market which their system is trading, then we can discuss the details as to why and how to trade such and such market. In my opinion, the market is very unpredictable and very chaotic, especially as we move towards more and more electronic base trading systems. Therefore, the current market conditions create chaotic behaviors in the market place which in turn translate into probabilities of profit/loss factor at anyone point in the market place. Sound like the Heisenberg principle?
Heisenber theory:
"Heisenberg took this one step further: he challenged the notion of simple causality in nature, that every determinate cause in nature is followed by the resulting effect. Translated into “classical physics,” this had meant that the future motion of a particle could be exactly predicted, or “determined,” from a knowledge of its present position and momentum and all of the forces acting upon it. The uncertainty principle denies this, Heisenberg declared, because one cannot know the precise position and momentum of a particle at a given instant, so its future cannot be determined. One cannot calculate the precise future motion of a particle, but only a range of possibilities for the future motion of the particle. (However, the probabilities of each motion, and the distribution of many particles following these motions, could be calculated exactly from Schrödinger’s wave equation.)"
This is the part what I like the most:
“The uncertainty principle denies this, Heisenberg declared, because one cannot know the precise position and momentum of a particle at a given instant, so its future cannot be determined. One cannot calculate the precise future motion of a particle, but only a range of possibilities for the future motion of the particle”
Does it sound like chaotic market dynamic and market instability anyone? If we believe that any introduction of a single entity into the market place which we cannot calculate the precise future heading then what we have here is 50/50 % chance that entity will travel up or down. Fortunately , the Heisenberg principle also states that “only a range of possibilities for the future motion of particle can be calculated” which mean that it is possible with the x-factors current market conditions , we can predict the range of that entity will end. So here , we have probability and distribution of entities that can be calculated which mean that if we believe in a trading system , the probability of success of a single system over time should reflect in the distributions of enties of that system. So let’s recap:
1) Heisenberg principle
2) Distributions of particles
3) The probabilities of each motion, and the distribution of many particles following these motions, could be calculated exactly from Schrödinger’s wave equation
System design:
1) Market movement
2) entry orders
3) probability Profit/loss of a system can be calculated over time because of many entry orders which will give distribution.
Remember , the ideas behind the Heisengberg principle is a tool to bring us back to reality that if the market price is a single particle travelling at high velocity it will be impossible to determine its future. However we can calculate the range of its motion. Therefore, we can structure our trading technique or system around that Heisengberg principle. So theorically , if we can control the distribution of our entities we may be able to trade the market after all. I think the important part in designing a trading system is to control the distributions of the entities, and in an unstable market it can be difficult but not impossible.
There is a little problem. While Heisemberg principle can apply to phisical reality, in the financial markets there are partecipants that ‘remember’ in some way what happened. Moreover they remember better the relative older facts rather than very recent.
Moreover again, they believe to be free but really they are following some common rules, common to the human mind. Otherwise there would be no explication about he fact that the bear market is more impulsive than the bullish one and that in the bear market the computers gain more money (it seems that bear markets are less erratic).
"There is a little problem. While Heisemberg principle can apply to phisical reality, in the financial markets there are partecipants that ‘remember’ in some way what happened. Moreover they remember better the relative older facts rather than very recent.
Moreover again, they believe to be free but really they are following some common rules, common to the human mind. Otherwise there would be no explication about he fact that the bear market is more impulsive than the bullish one and that in the bear market the computers gain more money (it seems that bear markets are less erratic). "
Yes , in a real market people are in control of their investment and trades. And yes, they based their decisions on historical data to project into future market. However each individual would remember x-facts, y-estimations, and z-solutions , therefore the possibilities are enormous and if you take that single individual multiply by an x-population of investors, now you are creating a chaotic, unpredictability, and instability in the market place or simply put it volatility. The uncertainty principle denies this, Heisenberg declared, because one cannot know the precise position and momentum of a particle at a given instant, so its future cannot be determined. One cannot calculate the precise future motion of a particle, but only a range of possibilities for the future motion of the particle. So in a market place if one cannot predict the sum of individual decisions then how can we predict where the future value of the market? Further more , each action in the market could theoretically bring the market any where in the trading matrix, such as range , out of range, break out, break down ect… So how can we assume that because of one single individual has analyzed x-historical data then surely most individual will follow the same path (trend trading anyone?) Well, it is true that in some case the path of x-group of investors could influence Y-group of investors to move in the same path, but again it does not prove that the market is predictable at a precise point into the future. The condition of price trend is simply a large distribution of entities migrating to a certain pricing range rather than a specific price. Therefore it is not possible to predict the exact price of the market at x-future period. And the reality is if one would know the exact price in the future and act upon it, by the time it gets there the market would probably affected it dramatically. Therefore, only the price range of distribution could be estimated as percentage of probabilities.
The distribution is not normal, as Mandelbrot demonstrates.
I will focus on the fact that rather the normal distribution is in the rules followed by the partecipant’s mind. Fear influences minds more than greed and for many reasons fear is around the corner in the bull markets
I have discovered the Hunter Memory: in a minutes chart, trace an Exponential Moving Average with length 55 and displaced 21 minutes on the right (future). It seems that prices cannot go too apart from such an average. This is an about sure rule. Perhaps other rules do exist… This is not to say that I have the crystal ball…
In the previous post: around the corner in the bear markets, not bull. EXCUSE ME.
A policeman pulls Heisenberg over for speeding on the autobahn.
He asks Heisenberg "Do you have any idea how fast you were going?"
He replies “Nope, but I can tell you where I’m at”
I always understood the uncertainty principle as just a mathematical consequence of the Schrödinger equation. Both impulse and position are waves, and they are related by a Fourier-transformation. I.e. they are related like wave length and frequency. A small wave length implies a high frequency. Similarly, small variance of the wave (small uncertainty of position) implies a large variance of the frequencies (large uncertainty of impulse). So they can’t be both small at the same time.
But perhaps I understood it wrong. Otherwise, I don’t see the relationship with the behavior of markets.
What surely understand is that in mechanics, although the behaviour of the individual particles has a certain unpredictability, the behavior of a large mass of particles usually loses that property and is again predictable by the classical laws of Newton. So if you equate individual actors in the market with particles, then your analogy would imply that, alhough these actors are unpredictable, the total market is very predictable.
I don’t believe that the market is usually predictable; but if it is, then the actors will become predictable too, in contradiction with the assumption that they are unpredictable. Moreover, the individual actors will become predictable in such a way that the total market will again be unpredictable.
So I would suggest that the relation between actors and markets is the opposite of quantum mechanics: While the individual actors may be predictable, their total is not predictable.
Recently , I try to look for money management trading on the internet and could not find any extensive written research on the topic. So I have decided to carry out my own research on dynamic money management rather than finding a perfect indicator. I also notice that the trading community emphasizes more on entry signals based on an indicator such as momentum , moving average ect… but there is almost no real research on dynamic money management. I think this is the main reason as to why certain systems shot up 100% then suddently crash back to its original point and sometimes lost even further.Therefore , the balance of money management and trading indicator should be extensively study and those ideas should be extensively express in the back testing data. But I think to find a silver bullet for money management is probably the same probability as finding a magic indicator. However, I am currently doing extensive research in the domain of dynamic money management. I hope any system designers out there would be willing to share some of their general concepts about this topic.
Thanh,
There was very useful disscussion in ET about MM.
(http://elitetrader.com/vb/showthread.php?s=07204dac0ffb8d95b074deb612854a8c&threadid=59325&perpage=6&pagenumber=1)
You can find a lot of good links and info there. Dynamic MM is a little bit different issue. I’m not sure that you ment it 
Can you elaborate what do you mean under “dynamic MM”?
Eu
The law of casuality states that entities are the cause of actions - not that every entity, of whatever sort, has a cause, but that every action does; and not that the cause of action is action, but the cause of action is entities.
Many commentators on Heisenberg’s Uncertainty Principle calim that, because we cannot at the same time specify fully the position and momentum of subatomic particles, their action is not entirely predictable, and that the law of causality therefore breaks down. This is a non sequitur, a switch from epistemology to metaphysics, or from knowledge to reality. Even if it were true that owing to a lack of information we could never exactly predict a subatomic event - and this is highly debatable - it would not show that, in reality, the event was causeless. The law of causality is an abstract principle; it does not by itself enable us to predict specific occurrences; it does not provide us with a knowledge of particular causes or measurements. Our ignorance of certain measurements, however, does not affect their reality or the consequent operation of nature.
Causality, is a fact independent of consciousness, whether God’s or Man’s. Order, lawfulness, regularity do not derive from a cosmic consciousness (as is claimed by the religious “argument from design”). Nor is causality merely a subjective form of thought that happens to govern the human mind (as in the Kantian approach). On the contrary, causality - for Objectivists, as for Aristotile - is a law inherent in being qua being. To be is to be something - and to be something is to act according to its nature.
Natural law is not a feature superimposed by some agency on an otherwise “chatoic” world; there is no possibility of such chaos. Nor is there any possibility of a “chance” event, if “chance” means an exception to causality. Cause and effect is not a metaphysical afterthought. It is not a fact that is theoretically dispensable. It is part of the fabric of reality as such.
One may no more ask; who is responsible for natural law (which amounts to asking: who caused casuality?) than one may ask: who created the universe? The answer to both questions is the same: existence exists (and its corrolary: only existence exists).
Dynamic money management , in theory would prevent a large drawn down, however, the down side of that is excessive fees involved in trading that technique. Therefore, a trader would have to balance between on going money management and extending its profit target. Hypothetically , if a signal is generated at x time on the chart randomly and later on x+1 time and so on … how would a trader manage the underline possitions to reach profitability at the end of all exit trades. Now , assuming all signals have the same chance of reaching profit target, how would a trader or can a trader just use money management skill to trade into positive territory? I understand that in real life certain signals are better and fundamental news can improve the profit target, but I am currious about from a pure mathematical standpoind if it would be possible to trade according to money management rather than pure probability of a signals. I hope I have clarify a little bit of my hypothesis.
Use stops. Trail stops.
If you don’t use stops sooner or later disaster will find you
(please see your own trade).
“BTO 60 @ESM6 1313.08 4/7/06 10:01 STC 60 1290.92 5/23/06 17:15 ($66,499) ($66,499)”
You can find dynamic money management at Six Sigma.
There are two ways to control trading. 1) using stops. 2) dynamic money management.
Neither is an absolute correct way. In other words, you don’t have to use stops or you don’t have to use dynamic money management to acheive success. It is the effective use of each that leads to success and it may involve either method exclusively or a combination of the two. Six Sigma is probably 80-90% dynamic management as opposed to using stops and does well. It has a good deal of gains since inception and is now consolidating gains as the dynamic process gets more in tune with the market. I am not a firm believer that one has to use stops to control trading, but that’s part of my process and I am sure it won’t fit with everyone. That’s why there is a choice of systems here.
I can attest though that dynamic money management can be extremely useful when done right.
Can your system be programmed into code so that it can be tested along with discretionary trading, let said 90% computer recommendation with 10% human involment?
>You can find dynamic money management at Six Sigma.
> There are two ways to control trading. 1) using stops. 2) dynamic money management.
I’m curious how you define “dynamic money management”. Stops can be “dynamic” in respect to ATR, etc. and they can be dynamic in respect to price (a default features in TS have allowed 100% programmable trailing stops at price X and % Y since the early 1990’s).
Is dynamic money management allowing a market to go against you
7% (like your QM trade) and then exiting @ MKT? How is that different
than a stop 7% below your entry?
How is a hedge really different than a stop? If you are long ES and the
trade is going against you and you short a similar amount of YM how is that different from being flat (i.e. being stopped out)? Yes, yes, on paper you still haven’t lost in the ES, but if the YM is a true hedge you can’t make anything either. I’ve known “hedgers” since the 1980’s. Sometimes they don’t want the customer to see a closed loss so they offset the loser with a back month spread or whatever. Often they end up just stuck in the trade unable to admit the loss and unable to pull the trigger on anything else. And then their trading gets immobilized. This is just my opinion: take your losses quick and put them behind you. Holding them and/or hedging them just extends the pain and the stress. Just ask Jonny Primetime.
BTW, speaking of “hedgers” check this out:
Midas (Prudent):
STC 104 LBU6 @ MKT GTC 9/15/06 9:59 9/15/06 9:59 277.9
The actual high for LBU6 on 9/15/06 was 273. So this fill benefits
the bottom line by $50K+ beyond the most optimistic fill possible…and FWIW total volume on 9/15/06 was 36 (good luck on STC 104 contracts on the last trading day). “Prudent” indeed.
And get this, C2 gives the fills an 85%/100% RF! Yikes!
Also:
BTO 52 USD/ISK 70.61000 8/28/06 17:39 STC 52 70.66000 9/13/06 9:56 ($915,200) V.High $367
BTO 295 LBU6 282.44 6/2/06 10:03 STC 295 281.81 9/15/06 9:59 ($352,464) High ($20,393)
So the max drawdown was somewhere between 915K and 1.2+ million
according to C2’s own trade records and yet no such dd shows up in the track record, and the “prudent” system is allowed to keep trading.
Since I don’t trade the Iceland Krona I’m not sure if the data is correct,
but SOMETHING is wrong.
Thanh
Dynamic money management , in theory would prevent a large drawn down, however, the down side of that is excessive fees involved in trading that technique.
Sounds as martingale for me lol, but yeah… you refer to not “dynamic MM”, as I understand, you refer to self-tune MM and it’s very different story. Any MM is dynamic by its nature.
You said you can code (Can your system be program into code). Pls, take my words as simplified example.
System: Flipping a coin.
Trading capital: 1000
Modifier 1: minus 2% from any bet. (The modifier supposes to make the game as negative sum game)
MM 1: No MM. We’re betting 1, per flip of coin. No matter what.
MM 2 (self-tuned):
Trigger: win/loss
Action(modifier 2): increase/decrease bet to 0.1. (e.g. if we won on previous bet, next bet will be 1.1. If we loose previous bet next bet is 0.9)
Code it You’ll see that the same system (flipping a coin) with MM1 is in permanent lost. The same system with MM2 is in loss as well (it’s practically, impossible to turn unprofitable system to profitable with MM) but the loss is very minimalistic in second case.
Lets take 10 coins. 8 are normal(.5 probability), and 2 are with defects that increase probability to 55% (0.55). The system is the same, but first MM still loose its bets, when second MM practically stops trading 8 coins and concentrates its capital only on two coins.
The example is very, very, very simplistic, but you can verify it yourself without trusting me.
Self-tuned MM is very young. Practically, it’s a system over trading system. I don’t think that you’ll be able to find anything in internet/books. And yeah… it’s expensive.
Eu
"I don’t think that you’ll be able to find anything in internet/books."
Google “Kelly criterion” for an introduction
