.

Chaos theory made simple - for trading

Moderators: mntiwana, mrtools

#51
Thank you Darkdoji for your reply.

Although I like the insights from  fractals and chaos as a mathematician, I haven't yet seen anything concrete for practical trading.

I respect Williams, and his "fractals" have some usefulness, but they have nothing to do fractals in Mathematics.

IFS are good for generating fractal objects (e g ferns, etc) but I doubt if the market is deterministic. By the way, the paper I mentioned before does not mention IFS.

Prof Didier Sornette is an expert on multifractals, I believe he is working on using multifractals to detect forthcoming market bubbles. This could be very useful.

Peters with his Fractal Market hypothesis (FMH) sees the market more like a stochastic process (not deterministic), but where the probability distributions are not normal. In FMH outliers are "normal".

I think Mandelbrot agrees that the market is random if you read the article on "the Seven states of randomness" (  https://en.wikipedia.org/wiki/Seven_states_of_randomness )

These are my views.

Best regards


#52
Teduh wrote:Thank you Darkdoji for your reply.

Although I like the insights from  fractals and chaos as a mathematician, I haven't yet seen anything concrete for practical trading.

I respect Williams, and his "fractals" have some usefulness, but they have nothing to do fractals in Mathematics.

IFS are good for generating fractal objects (e g ferns, etc) but I doubt if the market is deterministic. By the way, the paper I mentioned before does not mention IFS.

Prof Didier Sornette is an expert on multifractals, I believe he is working on using multifractals to detect forthcoming market bubbles. This could be very useful.

Peters with his Fractal Market hypothesis (FMH) sees the market more like a stochastic process (not deterministic), but where the probability distributions are not normal. In FMH outliers are "normal".

I think Mandelbrot agrees that the market is random if you read the article on "the Seven states of randomness" (  https://en.wikipedia.org/wiki/Seven_sta ... randomness )

These are my views.

Best regards
Well more thanks to you for coming in with what is clearly an informed perspective to initiate what would in the end be an interesting discussion. If you go to forexfactory now you find nearly the entire retail setup in open outcry as a result of the market reaction to FOMC releases yesterday. Let me use that to demonstrate how this thing works without being too precise (since even if I gave the matrix on which the market may be calculated to all here, there will still be doubts expressed so only money (returns from live trading) solves this thing). In the following post I discuss some of the points you raise here more directly.  Now this outcry against market outcomes simply proves that all these years no one in the prop market has understood how price moves. It is not just traders (retail or otherwise) it is also pundits (from banks, the financial press, etc) and all others who lay claim to understanding how the markets work. Observe that as incredible as it is, everything is speculation in this business despite years of existence. For instance someone or a group of players got it right with the FMOC (by error or design). Price moved against the rest of you, so someone or a group (organized or not) took the other side. But even that is an unknown - who took the other side and why? So nearly all known theories about how the market works do not work. I have said it often that it is wrong to think the market is an economics laboratory set up to trade delta in economic variables. Ultimately the way the market is setup allows every participating entity to trade or satisfy it's understanding of the balance of forces at different points and the market simply aggregates all incomings to sign the flow (so whatever strategy you use is bound to be right sometimes even when there is just a single explanation or pattern for how the market works). That should explain why so-called fundamental analysis is no different really than reading the market by the phases of the moon. The true solution of course is adopting the precepts of chaos theory and fractal geometry where you calculate market direction and range by significant turn factors called iterates or spot pivots. The Euro for instance and by that method fell to 1,0340 and was iterated to the upside at that point across all market scales. To the left of TF (or the unit) this low was reiterated at 1,04935 and this was well before the FOMC. Therefore, even on a probabilistic basis the Euro was and is rising (given the scaled iteration and range in terms of space in direction). This pattern of iterations (related to scale) thus suggests a turning level (down) at 1, 11268 (the attractor level broadly speaking at which we recalculate the market). This of course suits the medium-term trader as a goal but the fluctuations to that point remain complex and unpredictable because it is a chaotic system we trade (and so there is no point in predicting the course of price even with our understanding of the sequence of iterations to date. We need to just follow the matrix of iterations). However, this view (of a rising Euro) is fundamentally supported by the flow type observed for the Euro across scales and that is the fact of persistent waves at the lead scales - very simple (and almost sure therefore of our casing of the market). Similarly, in Gold we see an upside from a low attained before the FOMC but the suggestion reading the market by chaos is that this is a needed reaction up on the scale to persist down in the current unitary phase. Why? Because for now the perturbations are clearly localized within the overall flow (current unitary phase) and do not appear strong enough to upset the sequence of iteration and reiteration at 1375.08 and 1263.82 respectively. So to near certainty Gold was always going down. But we do not depend on predictions to trade the market rather on a matrix that describes chaotic behavior in market space. This means of course that regardless of the important knowledge we have on especially Gold (since it will accelerate 1122.68 into a sustained down phase on a unitary level when it gets there) we trade the immediate sequence which is now at or nearly about its return line by our calculations. When the market is populated by entirely separate entities pursuing their self-interest we must depend on the market and the market alone to tell what is going on - and that is accurate only when the market is read by its fundamental structure which is chaotic and fractal in expression. In the following post I will address more directly your points above - this is just to show you (hopefully) that far from being the complexity you may think this is (or even that the suggested application is unfounded) my desire is to indicate that there is a pattern dictated by the very structures we explain in "the chaology of markets" that is simple to use and that reads this complexity into direction and range - though it needs some pretty expensive gear to deploy - gear which even I does not have yet but is arranging to get - which is why I am out here talking instead of showing by result that this is settled). In any case, as a mathematician. I hope you can see that I am not misleading the public in my assertions above and that the market is a nested system of pivots (iterates) that can always be depended on (when correctly ordered per scale) to show what the play is and that as such even when a trader is wrong initially the trader has enough information to quickly correct his/her position in the same flow to profit. Cheers

#53
Teduh wrote:Thank you Darkdoji for your reply.

Although I like the insights from  fractals and chaos as a mathematician, I haven't yet seen anything concrete for practical trading.

I respect Williams, and his "fractals" have some usefulness, but they have nothing to do fractals in Mathematics.

IFS are good for generating fractal objects (e g ferns, etc) but I doubt if the market is deterministic. By the way, the paper I mentioned before does not mention IFS.

Prof Didier Sornette is an expert on multifractals, I believe he is working on using multifractals to detect forthcoming market bubbles. This could be very useful.

Peters with his Fractal Market hypothesis (FMH) sees the market more like a stochastic process (not deterministic), but where the probability distributions are not normal. In FMH outliers are "normal".

I think Mandelbrot agrees that the market is random if you read the article on "the Seven states of randomness" (  https://en.wikipedia.org/wiki/Seven_sta ... randomness )

These are my views.

Best regards
Now to respond more directly to the points you make here.

I take your first point to imply that you have not read my presentation "the chaology of markets" before arriving at that conclusion. If on the other hand you have (and I believe we did a clear job of stating exactly how the market works by chaos and by fractals) I shall be grateful (for a mathematician as you are)  to point out where in that presentation I have adduced material that confirms your current position? Is there any point that was not covered to make the presentation not the joined-up thinking it was intended to be? I shall be most grateful and enlightened by your views.I have no regard for what Bill Williams thinks and consider him strictly a technical analyst (regardless of whatever so far unexpressed knowledge of chaos and fractals he may have). It is this point you made that indicates to me that you may not have read my output yet. For a mathematician to associate even remotely my ideas with Bill Williams (after reading what I have had to say in the chaology) is simply not possible. As such I can only assume you are speaking without reference to my stated position. Please read the material if you do not mind.Again here you miss the point. What is a bubble? How much of that is your concern from day - day as a trader? I do not doubt the expertise you claim for Prof Didier Sornette and I wish him all the luck in the world. But frankly, the topics are different and whatever the good prof. comes up with will have to be evaluated with the basic setup of markets which we have declared in the chaology indisputably (unless of course you can guide us as to where our math is wrong or misleading). Do not forget this is where I think Mandelbrot's profound knowledge as he has shared, is misplaced; trying to address quants and risk managers in the control of risk when in fact it is only by the trader (and his understanding of how the market works) that risk can be reduced. This is why we have provided a general multifractal trading model to any type of trader - algorithmic or otherwise to use at this time. Ultimately, fractals as it applies to the market is done and gone. Mandelbrot provided all we need to know and to the level that low level intellectuals like you and I can understand and deploy right now (I only hold 3 certificates of which the highest is a mere masters from some Scottish university). Bubbles do not matter much to me in the Gold (and sometimes currency markets) I care to trade. But I understand the sentiment - approval from academia is what we must live by and I do not argue with that - except to say that when that time comes (and if we are still alive) it will not dispute a dot of what we have described in the chaology.Again we can ignore (for the present purposes) Peters and his hypothesis. He like most academics and near academics have needs not shared by traders. As a mathematician I am sure you appreciate the extent to which mathematics is not arithmetic and as such you would know that much depends on the assumptions made whatever the math applied. For instance, in algorithmic trading, which I am very interested in as a prop trader, Glosten and Milgrom as extended by Samnay Das posit alpha to indicate the proportion of Informed Traders in a flow. These are traders who according to them know all future points all of the time and exactly (100% of the time). The rest of course are assumed to be uninformed traders who are as lost as the informed traders are gods in this setup. Now the evidence is clear that there is in fact no basis in reality to make such assumptions except for the simple convenience of making the ensuing calculation "portable" within the mathematical formalisms adopted for the calculations. GS, CiTi, Saxo or anyone else does not fit the bill of the know it all trader nor do the profits of such assumed characters (in real life) reflect that reality. But there are thousands of trades made based on exactly that thinking (baseless as it is) and this reflects (not surprisingly) exactly - the thinking of the pagan trader who assumes the market to be fixed. But what is worse, although the proportion of informed and uninformed traders in the market is assumed to be known, no details are given about how to calculate this property (so as to even glean the realism in that assumption). Strange is it not? Then inter alia, the model goes on to distribute probabilities according to the normal curve (and all that is implied by that for how price moves). So we have the sort of rigor I was taught to get a masters but so what? The model fails and does so badly (as in fact it trades with no more understanding than we human traders have about the market). My point is lets not mix it up. The trader needs to understand how price moves in real time to calculate a response. Mandelbrot already provided that and all I have done is to show how it would work in electronically traded markets - if we find that true then we can always measure what others are saying against such an understanding (Peters in his hypothesis introduces an assumption so similar to the one we just discussed that I worry about. Mainly because everyone (except Mandelbrot) just tries to avoid the market in terms of point displacement as the mechanism to understand which is what fractal geometry is setup to do. And that is how traders can read the markets, so it is fair to say Peters is not talking to traders and so we can ignore peters (I have) on that score as traders).IFS is mentioned in the paper you highlighted and very elaborately too (i.e. to detail). The whole point of those elaborate diagrams Mandelbrot was so eager to show us all is IFS. You being a mathematician can easily check this with any professor of your choice anywhere on earth. I take it that you have not had more than a passing interest in fractal geometry not to know that. But please correct me if I am wrong; in the chaology, we show you how and why the IFS is pertinent in any approach that seeks to understand how price moves. It should be easy then to dispute my findings simply by showing that what I describe cannot be. But if you imply "not mentioned by name" then consider the book "The (Mis) Behavior" of markets by Mandelbrot and another, where the same material is represented and you should then have no doubts. Finally, Mandelbrot CANNOT disclaim his own thesis can he? You say elsewhere that "where the probability distributions are not normal" and acquiesce or at least do not dispute that assertion, then how can Mandelbrot "think" as you assert? Fractals are distinguished by being deterministic variables that appear random in chaotic space and we make clear why that is in the chaology. Again being a mathematician it is more than simple for you to affirm this with a relevant math professor anywhere on earth. With regard to that point then - I can only say, I think you are very wrong and may have looked at the material you cite - only sparingly.So all in all - those are my responses and I am happy to clarify any areas not immediately clear (I am sorry my English is completely inadequate so there may be areas that would not be clear to a native speaker if you are). In any case, I hope you will respond so that we can continue this already very interesting discussion. Thanks so much for your time (why? Because you help to validate my ideas and prove its worth to all interested in breaking the gridlock we have had in trading for so long).

#54
Hi,

I am sorry, I don't want to be drawn to a long winded discussion due of my time restrictions.

Let me just clarify that I am a mathematician, but also a software engineer and a trader for over 10 years. I have much respect for all the people in different areas. When I say I value Williams ideas, that was me as a trader. When I say his notion of "fractals" is not mathematical fractal, that was me as a mathematician.

In general, I remember the English saying, "the proof of the pudding is in eating". Mathematicians must prove their beautiful intuitions. Computational ideas must be implemented as programs that works. Trading ideas must show results. That why I am in this forum because I know mladen and friends do a very good work in programming trading ideas.

All the best

#55
Teduh wrote:Hi,

I am sorry, I don't want to be drawn to a long winded discussion due of my time restrictions.

Let me just clarify that I am a mathematician, but also a software engineer and a trader for over 10 years. I have much respect for all the people in different areas. When I say I value Williams ideas, that was me as a trader. When I say his notion of "fractals" is not mathematical fractal, that was me as a mathematician.

In general, I remember the English saying, "the proof of the pudding is in eating". Mathematicians must prove their beautiful intuitions. Computational ideas must be implemented as programs that works. Trading ideas must show results. That why I am in this forum because I know mladen and friends do a very good work in programming trading ideas.

All the best
Of course, but this is a thread of record and it is important that we clarify especially the points made by you that do not serve the cause of knowledge since it is a public thread and others will come after us. There is no need to be apologetic about your interest in and use of Bill Williams ideas. He is in the ideas market and what he sold you bought and use (at least some quantity of it) and you have a right to do whatever (to your mind) facilitates income growth trading. However, I agree with you absolutely about Mladen and friends - they are fantastic coders and so we are attracted by the same factor or similar factors to this site. In fact Mladen was among the very first group of three persons I sent my presentations so you must know that I respect him for his person, coding ability and his math. He did not acknowledge that dispatch before I came here and of course I do not know whether he actually saw it before now because he and the crew were no longer at the venue to which I dispatched. Indeed, long ago when I wanted to code some of this stuff (that time still in its most raw form) it was to Mladen I turned (with encouragement from mr. tools) and he asserted and maintained he would work on tools only discussed in the forums (Forex-TSD). So I have used Russians on a commercial basis since. I am here to keep with what is going on with "technical" indicators from the masters of that game. But lets not be deluded it is NOT about indicators - or we would have found the ultimate by now and reversed the fortunes of especially retail traders which is clearly not the case to this point. It is about reading the market by the correct sense of the market and building indicators that help that sense - fortunately "technical" analysis is the "poor" cousin of chaos and those tools apply even when the context is chaos. Just that the sense in which they do changes. This means no chaotist will touch harmonic butterflies and Elliot Waves, or the 9 of squares of Gann, etc. As to proof we share the same sentiment exactly. I have merely put this out because I may not be around beyond a point (you do not know my exact circumstances) and wanted to be sure that enough was said to assist anyone interested to find the same meaning as me just in case. So thank you all the same for allowing me once again to validate my ideas. Thank you very much as I am especially glad that you are a mathematician and trader and now a technical person as well. Best wishes in your future work.


#56
Hi,
i'm loving this idea. The problem is implement it.
Regards,

#57
MBeng wrote:Hi,
i'm loving this idea. The problem is implement it.
Regards,
The idea is simple and very beautiful and it is also incredible that we have remained blind (especially in retail trading) for so long. Do you know that brokers are now advertising to get retail people to trade algorithmic systems (i.e. you forget about trading yourself and simply subscribe to such a service?). Implementation is a problem only because a) The equipment is expensive (as we specify it, the console alone plus shipping to my country for instance is $8.5k though folks in the USA will get lower shipping costs) then of course you would need space etc. b) because I am the one who cracked it you are gonna have to depend on my setup initially which of course is in line with what you have read and only after a community evolves are we going to get ideas from others etc c) It is untested - i.e. the whole scheme is not tested in live trading yet so in theory it is a significant risk (though I am pretty confident but that confidence has to do with years of working on it and knowing things about it that is not immediately obvious to others). But we will see how it goes. I shall, once I am setup and trading, be in a position to start posting details and in such a way that no one is misled who is interested. But I can almost assure you - it is a revolution - and the laugh is that it took a trader questioning all the hopeless lies about how the market works and losing something like $65k in the process to call the absurdity of technical analysis and so-called fundamental analysis. But we shall see - I hope to be the guinea pig but open enough for others to emulate if they want. 

Cheers,

#58
So, how do you define  a fractal? How do you determine the dimension of fractal that will be replied? I undestand the teory but i can't make the bridge to markets. Can you elaborate a litle more?
Regards

#59
MBeng wrote:So, how do you define  a fractal? How do you determine the dimension of fractal that will be replied? I undestand the teory but i can't make the bridge to markets. Can you elaborate a litle more?
Regards
Sure, fractals are the price points you see everyday, a fractal is any three price turns following each other in the same flow on any scale at all. But understood for its properties we focus on its dynamics (not the shapes and at least not initially) and particularly the fact that fractals pervade the entire trading space. So in order to trade you need to understand the market as driven by fractals large and small in fact to the tiniest bit of transaction and to the largest deal available. This is not obvious when price moves because bars as an indicator of price hide  a lot (graphics matter). In any case, you use the IFS understanding to emulate price in market time. So we talk about fractal primitives in the chaology. So basically you have to calibrate space according to specifications that meet some standard that is easily verified. That is what I have figured out. To share the same understanding read the things I have read which is anything and everything about chaos and fractal geometry - you do not have to be clever just passionate - to know. Then I think the right questions will get asked and incredible justifications will be made - because then it becomes dead obvious. We can agree. So that is another power of and or attraction to the fractal reality that is good. The butterfly effect is your friend - a standard measure can be agreed. I am saying such that if I say the market is at some specified low in a given range everybody agrees (can read it objectively). Communication will take a turn for the better among traders and pundits and the financial press etc (you can't just shoot off your mouth). Fractals are "natural" algorithmic trades - whether you are human or machine is not important. So with a setup such that you can a) define a goal for a move that you read and time to inception 90% of the time. b) given direction of a move know all the local limits (e.g. across some scale) and what pivots at certain points imply (not because I say so but based on the intrinsic algorithm of the market) you will find the market far less confusing than is the experience now and therefore a lot more (far more) profitable. I am saying this such that in every trade we are trying to see how the market is scaling to a shape (usually a point at some scale - e.g. H4, D1 etc) that defines a triangle with time on the horizontal axis) just like in the chaos game. Have you played the chaos game? Please play it. I am sorry but you got to feel this thing it is there in the dynamics you see on your screen. Sorry I forgot to say about dimension we do not worry much about it. Roughly it means type of dynamic which is easily read in the Hurst types (so long as it is understood that in general, trading ranges for optimal decision making is too short for statistical methods (which just says we can judge things by precise math. But traders don't do statistics they do dynamics - the time to decide is just too short for traders).

.
PS: The idea by John Ehler and others to transform fractal dimension into some kind of indicator - rough (range) not so rough (trend) is actually laughable and very typical of technical analysis. It is a very shallow view of what fractal dimension is mathematically as in fact you know any oscillator already captures that exactly (rough and not so rough). Trying to make specific indicators based on that is very shallow indeed (which is the problem in general with technical analysis - total ignorance of how the dynamic must be measured). What we do is to define the price curve differently and interrogate this curve for a mix of fluctuation types across all scales and interpret the indications for direction and range based on the space in direction to the rightmost scale or time frame. 

#60
What I am saying really is that this idea of chaos and fractals does not reduce to an indicator or some system like Elliot Waves. It is a way of understanding how the market works and the only way is to first understand the concepts and how they apply to the markets generally - then we can talk specifics. We tried to outline the general picture in the chaology and it is questions from those thoughts that are more meaningfully answered and related to the market to allow the sort of understanding we all seek.



PS: I am not sure that the concepts no matter how simplified will ever be so simple as to read stuff today and apply tomorrow. This is serious business trying to understand these concepts - as in it requires some attention to details and thinking about what it means in terms of the market.


Who is online

Users browsing this forum: CommonCrawl [Bot] and 1 guest