Determinism always leads to disaster...

@Alex....Do you have a sample of your solutions' output signals for the various financial instruments it caters for, at various time frames??

I agree with Alex in that there's nothing random in the market and every tick has its meaning, it is what I've been advocating in this group for many years. At the same time I do agree with Valerii that this is not a sufficient condition for success. The point is that 99% of this meaning is unknown to anyone. However there is that 1% left that often provides statistical advantage that can quite successfully be exploited.

Hi Alex,

"Earning capacity of the trading strategy based on this methodology exceeds the industry benchmarks."

Can you provide some statistics? For example, I would be interested in knowing what is ratio Profit/MaxDrawn generated by your model for a period of 2 years.

Thank you

Well, based on my study of different approaches to cycle analysis, I would not be surprised if you were largely right.

Markets are a subtle combination of apparent chaos and intrinsic order - just like the universe actually. "Cracking the code" requires an open mind and a fair amount of time.

If you are brainwashed to believe in EMH, random walk, CAPM and all this pseudo-scientific nonsense spawned by mainstream financial academia, you will go nowhere obviously.

Alex, if you have any summary of your work - without betraying your trade secrets - to document your findings, I would be delighted to share views.

A theory of everything always leads to disaster...

Does this reconcile with the statement "the market can do absolutely anything"? Maybe it does, it is an honest question, not a snarky one :)

Actually similar to Newton's opinion about the possibility to calculate the future by knowing every single atom movement in the universe (thanks to Heisenberg's discovery, this doesn't work), the market movement could also be calculated in advance. We'd only need to know every decision logic (or gut feeling) of any market participant, his or her financial and technical capabilities, the network bandwidths and latencies (incl. all other hosts' future network usage behaviour - even of botnets or MP video games), and of course the exchanges' order processing and price fixing logic. But now we're even unable to predict upfront, whether a fund will make a decision act in a big way beginning at 10.37 a.m. (right after an internal meeting) or 2.00 p.m. as they usually do.

Seen a bit more realistically, we just have to optimize the scopes and views of the way how we watch and analyze the market. A candle of any time frame is actually a rather arbitrary view, but since it's used by many traders, it works somehow. Autocorrelation analysis as presented in many papers seems to be of limited use. Those values are usually calculated over a longer timespan and thus average out any short but special autocorrelation phases found in the data. All this can be automated BTW.

@Sofistar: A theory of everything (TOE) is still missing, so until now there was no opportunity to study its outcome. ;)

"Market is not random, it's highly organized, structured and governed by certain strict rules and restrictions;" That's probably true ... but there is a pecking order:

The House

Fools

Fool (er)

Fool (est)

and anything in between.

this is just a "conversation" (aka unfounded statements) until ......

you start publishing predictions over the course of e.g. the next 1 month. We verify them and calculate the residual error. So if your model(s) are so great then we will verify that ourselves, WHILE the confidentiality of your super models remains intact

ps: note to you and some of the other guys - here I am referring to predictions NOT performance stats about your model

Matthias Waldhauer: "...Newton's opinion about the possibility to calculate the future by knowing every single atom movement in the universe..."

To be historically objective we better say not "Newton's opinion" but "Laplace's determinism". Pierre-Simon Laplace wrote in "A Philosophical Essay on Probabilities" (English translation):

"We may regard the present state of the universe as the effect of its past and the cause of its future. An intellect which at a certain moment would know all forces that set nature in motion, and all positions of all items of which nature is composed, if this intellect were also vast enough to submit these data to analysis, it would embrace in a single formula the movements of the greatest bodies of the universe and those of the tiniest atom; for such an intellect nothing would be uncertain and the future just like the past would be present before its eyes."

Sir Isaac Newton, a famous natural scientist, philosopher and ... trader, is attributed other words after his unsuccessful trading the South Sea Company stock since creation of the company in 1711 until the bubble in 1720: "a man can calculate the movements of the stars but not the madness of men".

As for "Heisenberg's discovery", then unpredictable consequences can appear not only due to "randomness" (I put double quotation signs to emphasize that a definition of "random" was not yet given here). Newton's mechanics as it was shown by Henri Poincare faces serious difficulties, even, when everything is deterministic. Poincare's fundamental works on celestial mechanics were the first seeds of the modern theory of chaos and dynamical systems. If minor changes in the initial parameters of a dynamical system cause significant and fast growing difference in possible directions of its evolution, then the task of predictions also becomes notoriously complex, even, formally no "randomness" is involved.

Best Regards,

Valerii

Evo Eftimov: "... you start publishing predictions over the course of e.g. the next 1 month ...".

This would be excessive. If a system successfully predicts, then there is a serious alternative to advertising it: be silent, trade, and accumulate profits as long as it remains successful. An argument that advertising is done in order to attract capital to multiply profits is not sound because reinvestment of profits starting with smallest possible capital creates exponentially growing account curve. In other words: with successful predictions it does not matter when to start or how small an initial capital is (just enough for a margin required to trade at the beginning a single contract plus commissions) - becoming a millionaire within a year would be inevitable.

Evo: "I am referring to predictions NOT performance stats about your model".

Naturally, one should not at least begin asking to apply methods of the probability theory or statistics to something already claimed to be "non-random".

Best Regards,

Valerii

Agree with Valerii... If it works, stay silent and make money!

Alex

Velerii I am sorry but to understand what i said / suggested you need to be a) a Trader rather than a Risk Manager and b) think/live/exist in the real world

Evo,

Please, reread my message. It completely supports yours.

Best Regards,

Valerii

@Antoine - Alex doesn't necessarily have to stay "silent" to make money. Going "public" about the externally visible properties of his model can bring him even more money than starting to grow the equity curve from 0 / with his own money. Many hedge fund managers and traders are on CNBC every day precisely for that reason.

The key here is that "wonder" models have come and gone without living anything behind except "boasting".

Also Alex may want to "go public" for pure scientific glory - and that would be admirable too.

However like some of the "wonder" academic papers without any e.g. accompanying R code to validate / verify them independently, if Alex doesn't publish any predictions intraday / daily / weekly etc which we can verify independently, then for me he would be just the next guy who thinks markets have systemic rather than completely undeterminstic properties - and that is true - however coming up with a practical and effective model which can consistently take advantage of the systemic properties is not trivial and the only way is to show us some predictions on FUTURE data rather than back tested result and performance stats calculated on them in the privacy of his office

"Private group. To request membership, click Join ...".

I would like to join but I cannot find the button to click!

Evo, while I enjoy reading your pamphlets I think you don't need to be so harsh on Alex or other determinists: I used to criticize this approach no lighter than you do now, but then I came across seasonality. And then many things started to strike resonant chords in my mind.

Again, I don't intend to support the topic starter at least because apart from the very idea that markets are not random there is unfortunately zero information in his original post, and he doesn't seem to appear in the discussion.

Markets are neither totally deterministic / systemic nor totally nondeterministic / random.

(the only more or less persistent systemic relationships (and without intermingled with too much noise) can be found at market plumbing / market microstructure level.)

Secondly, the ratio between systemic and non-deterministic behavior vary constantly

And finally, whatever systemic behavior there is, it keeps changing over time.

Hence the ratio and nature of systemic behavior needs to be monitored all the time and the models adjusted accordingly

here is how we did it when invented a new model - it got us street cred and clients http://earlyannouncement.com/prior-nowcasts

Kristop, my friend, I do not "criticize" Alex - he was merely invited to back his claims with proof

Alex Krishtop makes some very valid points. Every tick does in fact have a meaning, we just can't always determine when or why these ticks happen. I try to stay as up-to-date with Forex market movements as I can (i.e. watching the news, reading business and economy and following groups very similar to this one). This along with my Fibonacci sequence help me to place consistently successful trades.

Jared,

Have you ever considered applying Predictive Analytics to Fibonacci concepts?

"Predictive Analytics about Fibonacci concepts" - that is interesting as a separate "concept" in itself ....

Fibonacci sequences already "predict" something (ie they are also predictive) and one of the reasons is that every man and his dog uses them .... (and many times this is used by the big guys to take advantage of the man and his dog)

So what exactly are we predicting about "predictive" fib sequences:

1. In what circumstances the big guys will eat the small man and his dog who used some fib analytics to make some trades ?

2. Fusing fib sequences / predictions along different dimensions ie identifying their intersection / confluence which is more predictive than the parts?

3. Both 1 and 2

The question to which predictive analytics (modeling/machine learning) could be applied would be to figure out why Fibs seem to precisely time significant turns, sometimes, and at others times seem to be unrelated to significant reversals. Now I will grant that a dedicated believer can always find a "resonant" ratios associated in some way with major market turns after the fact. However this exercise often requires ignoring things that just don't fit into the paradigm. Does the market go chaotic at these points only to return to resonance in a few weeks? If so why,how and how can this be predicted in advance?.......some food for machine learning..

for the protocol, "predictive analytics" is not only "machine learning / AI" and other sexy acronyms / catch phrases - the humble Histogram, regression and yes Fibs are also in the class of predictive analytics

re what happens in the cases when Fibs don't work for every man and his dog - well check point 1 in my previous comment - i am afraid it is not only nice /clean / scientific stuff such as a system entering a chaotic state / mode and hence Fibs (and other "nice and simple" models fed to the retail public) stop working suddenly (and again don't forget that Fibs is a self fulfilling joke in most of the cases anyway).

If you are a predator and know that the wildebeest goes to a watering hole called Fibs and where that hole is on the map you are going to take advantage of that

Yep, I hear HBSC has over 200 Fibonacci analysts sitting there all day with their protractors and charts plotting how to rob pension funds and small investors....but it's a secret so let's not tell the public.

Obviously you are very well connected mate

Btw you got the "robbing" thing slightly off the mark - nobody is robbing anybody - it is a game / business called market - and anybody who doesn't "train" and get equipped for that game properly, will loose - you would not say that whoever won the 100 meter sprint at the Olympic games "robbed" from gold medal the couch potato watching the games on the TV while drinking beer, would you

AlI I know about the plots to control and manipulate the markets I read on the Internet. All the secret plots are posted there and the Internet never lies so we know it is all true. I'm sorry that the big guys have taken such advantage of you and your dog. This is so sad.

this thread is full with people who do not take time to reflect and shoot off the hip without thinking - there are no "plots" mate - it is a game / competition / business, where everyone is given a chance to compete under the same rules. People who use simplistic (in many cases obsolete and now naïve) analytics will be taken advantage of ie they will keep loosing the competition

better get trading for real than "reading the internet"

You and your dog be safe now!

there you go - ran out of logical, fact based arguments and resorted to personal jostling - what an utter, amateur crap