From theory to practice - page 1489
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
This "undoubtedly legitimate existence" is not one without mathematical proof.
Unfortunately.
And there's a lot of questions, such as:
"how did you determine that existence, if any, is a pattern?"
"what is the kind of trend?"
"why are they different?"
"where did the concept of them alternating come from?"
"on what basis would a profit and does it then take into account the existence of spreads, commissions and swaps?"
Are you prepared to answer these questions reasonably and prove that this is in fact the case?
And it turns out that practically any "author's" theory on this forum, being put to the test of such questions will simply turn out to be untenable.
In order to prove something, one must first make small, microscopic, but certainly reliable mathematically grounded and calculated steps in order to prove something based on this foundation. If anything can be proven based on such a foundation at all.
You should not "prove" here, but continue to make your (possibly unique) TSS trend-tracking ("trading") system
Any explanations in private (if you really need them)
Is there even one person who can mathematically prove at least one constant and unchanging market trend that can be mathematically calculated by anyone and make sure that it (the trend) really exists? (And do not tell me that you have super-secret developments that no one needs to know about and that they work somewhere - this is paranoia bordering on madness).
Otherwise, it feels like there's a forum of mostly "unrecognized geniuses", frankly...
And besidesAlexander_K... there's no one here who's even sane...
But even he for the most part believes in some kind of witches and the like, which in general can not help but cause alarm, but nevertheless at least try to scientifically and importantly, reasonably talk about methods that can really work and always.
I'm not trying to make myself look like Dantes. No. And I don't want to offend anyone. I'm just stating the facts as they are, drawing another line.
The basic law of the market is that any pattern that allows you to make a profit is short-lived. This is quite reminiscent of the famous Winnie the Pooh song about a strange object - honey. Mathematically it can be expressed through models of mathematical game theory, but it makes no practical sense (for trading).
You should not "prove" here, but continue to make your (possibly unique) TSS trend following ("trading") system
Any explanations in private (if you really need them)
A "TSS trend following ("trading") system" - does it imply an analysis of the market situation or a specific order opening sequence?
Good, excellent questions, Che. Why is it that there is no unified concept of understanding the market and everyone tries to overcome it to the extent of their education, within their world view and is categorically unwilling to listen to others?
Here is my opinion on the matter.
I have done probably a billion studies on the topic - is the SB the market or not? And this is precisely where the main problem lies.
The most appropriate random process to describe the market is Laplace motion. So what do we see? All the central moments can differ by orders of magnitude at a certain point in time, and coincide at another point in time.
You gave figures once: the market is 98% SB, and 2% is laplace motion. I'm afraid the real numbers are different: 75/25 or even 50/50.
Alas, we are dealing with a random process with memory, a non-Markovian process. Moreover, "memory", i.e. dependence of sequential price values on each other forming simply gigantic, inconceivable from the point of view of the random process theory, trends, does not always exist, it appears episodically.
In short, we have a process poorly studied in terms of modern physics and mathematics. So what to do? That is why everyone works as hard as he or she can.
One cannot apply to the market the methods of struggle, which, ideally, can even defeat SB (such as Martingale or Warlock indicator, based on the convergence of the sum of independent random variables to the Gaussian distribution) or formulae from the linear equations of motion, as the Automat does.
The market, like a two-faced Janus, will tear the two strategies apart. Therefore, a really working strategy must combine both approaches - random and deterministic. And this is difficult, very difficult.
P.S. It's been a long time since I've written such a scribble, but Che's questions are worth it.
Good, excellent questions, Che. Why is it that there is no unified concept of understanding the market and everyone tries to overcome it to the extent of their education, within their world view and categorically does not want to listen to others?
Here is my opinion on the matter.
I have done probably a billion studies on the topic - is the SB the market or not? And this is precisely where the main problem lies.
The most appropriate random process to describe the market is Laplace motion. So what do we see? All the central moments can differ by orders of magnitude at a certain point in time, and coincide at another point in time.
You gave figures once: the market is 98% SB, and 2% is laplace motion. I'm afraid the real numbers are different: 75/25 or even 50/50.
Alas, we are dealing with a random process with memory, a non-Markovian process. Moreover, "memory", i.e. dependence of sequential price values on each other forming simply gigantic, inconceivable from the point of view of the theory of random processes, trends, is not always present, but appears episodically.
In short, we have a process poorly studied in terms of modern physics and mathematics. So what to do? That is why everyone works as hard as he or she can.
One cannot apply to the market the methods of struggle, which, ideally, can even defeat SB (such as Martingale or Warlock indicator, based on the convergence of the sum of independent random variables to the Gaussian distribution) or formulae from the linear equations of motion, as the Automat does.
The market, like a two-faced Janus, will tear the two strategies apart. Therefore, a really working strategy must combine both approaches - random and deterministic. And this is difficult, very difficult.
P.S. It's been a long time since I wrote such scribbles, but Che's questions are worth it.
"TSS trend following ("trading") system" - does it imply an analysis of the market situation or a feature of the order opening sequence?
And also, for example, the presence of the following elements, concepts and definitions:
trends local, composite, previous, pre-existing, current, visible, hidden;
bars previous, current, paired, their properties and states;
events, states, reversal, breakeven, continuation zones, specific points;
visualization of the next trend, control of cumulative volatility and return;
current and cumulative gains in current and hidden trends;
control of basic operations (B,S), movements (pullback, rollover, rise, reversal,
reversal, continuation, idle), locations (beginning, middle, end of a trend)
controlling volatility and profitability of trends and trades
optimizing lot sizes, opening and closing orders;
displaying (during debugging) the size of current trends and other necessary data.
Is this enough? Or is something missing?
And also, for example, having the following elements, concepts and definitions:
trends local, compound, previous, previous, current, visible, hidden;
bars previous, current, paired, their properties and states;
events, states, reversal, breakeven, continuation zones, special points;
visualization of the next trend, control of cumulative volatility and return;
current and cumulative gains in current and hidden trends;
control of basic operations (B,S), movements (pullback, rollover, rise, reversal,
reversal, continuation, idle move), locations (beginning, middle, end of a trend)
controlling volatility and profitability of trends and trades
optimizing lot sizes, opening and closing orders;
displaying (during debugging) the size of current trends and other necessary data.
Is this enough? Or is something missing?
Total :
Nothing will work at all, unfortunately, for an aty-two!
On the other hand, the devil is not as bad as it looks. In an already running programme with the above stuffing, which gives very good results, there are only a little over 160 lines
Continuing.
So, let's settle on the fact that the market has a 50/50 randomness/non-randomness property. Let this be a working hypothesis.
Starting this thread, it seemed to me that I could easily cope with the task - it is enough to convert BP to a random process and use the regularities of SB, namely, the constancy of the variance resulting from the Einstein-Smoluchowski equation, the return to the starting point in 66% of cases in two-dimensional SB, convergence of the sum of a large number of independent random variables to the Gaussian distribution, etc., to have a profit.
Alas, this turned out to be an unsolvable problem. No transformation of the original BP allows to bring it to a pure random process.
So, I had to start a long and tedious search for the KEY to randomness/non-randomness of the current state of the market. And this is the right approach to take.
Every trader needs to know - how exactly they can make money in the market under ideal conditions for them. The problem of making profit on some idealized model - random or non-random - should be solved unconditionally.
Once again, my model is a random Ornstein-Uhlenbeck process with the property of returning to the mean. I know how to capitalise on it.
So, the hardest, but achievable task is to be sure that all conditions of this model are fulfilled at the moment of entering the trade. For this purpose, my TS has as many as 4 (four) keys - parameters that assess proximity of the current market condition to this model.
Results, stats, signals - all this from September 1.
Now the most important thing - each trader must have a model on which he/she can earn and a key, defining whether the market satisfies this model here and now.
Everything.
Once again, my model, is an Ornstein-Uhlenbeck random process with the property of returning to the mean. I know how to make money on it.
I must have missed something....
what about the last $100 deposit - there were some screenshots, then up and not even a hint of successful trading and "ripping pockets" and "going to the factory"
So, the hardest, but achievable task is to be sure that all conditions of this model are fulfilled at the moment of entering the trade. For this purpose, my TS has as many as 4 (four) keys - parameters that assess how close the current state of the market is to this model.
It's "elementary, Watson! (C) - use StopLosses and you will immediately see everything!
And the key, which determines whether the market here and now satisfies this model.
is "Elementary, Watson!" (C) - use the MT4 /MT5 strategy tester