![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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
Used is Aleksandr?
because I'm already confused who is who...
Why would I be?
Volatility (RMS) is simply an estimate of the variance of a time series (sequence). If we calculate this estimate for each new member of the sequence (similar to how MA indicators are calculated), we can get a new sequence of volatility indicators, in other words, a new "derived" time series, which can be stationary (the mathematical expectation does not change with time).
Where did I get that from? Just so understood your words.
Suppose we have 8 7 5 2 (the combination is rare but don't mind it for clarity)
Let's build a sequence of incremental modulus differences.
|7-8|-|5-7|,|5-7|-|2-5|.... i.e. -1,-1,... It becomes more probable that the next number will be with + or 0. But plus can only be if the number 8 or 7 or 6 or 5 or 4 or 3 comes out
The "sequence of incremental modulus differences" may be a stretch to "volatility sequence" (as opposed to RMS, people often think of volatility as something of their own)." It becomes more likely that the next number will be " interpreted as a reliance on stationarity, on the return of the series (the tendency of the next sequence value) to its MO.
If I understood it wrong, I'm sorry.
P.S. The offended will be punished.)
So far I see, combinatorics is at work (a method of analysis I haven't yet encountered).
Probably the obvious question for those who have understood something: on what basis do you choose the depth of history of combinations in the past? I think it all depends on that in this case...
For example you have chosen combinations of 8 5 7 2. But in the past for example there was another outcome of three eagles, in binary calculus equal to 4, then the combinatorial analysis is subject to 4 [8 5 7 2] and it can change the set of resulting samples dramatically.
As for the increments from the last post.
Paragraph1
Here you have assigned all possible combinations of 3 heads and tails (there will be 8)
Numbers from 1-8 (arbitrary!). You have a sequence of numbers from 1-8.
ITEM 2
Next, you must choose the conditions under which will be a kind of stat advantage (as here for example: And now let's remember about the price of the game. And we will remember it quite simply. Take, for example, the dogma "A game is profitable if the probability of winning exceeds the probability of losing by at least 2 times". I.e. with respect to our dynamic probability from the example it means that one should start to bet on black, when the dynamic probability of black winning is at least twice as big as the dynamic probability of red winning (VHF >= VAC). For our roulette example you have to bet on black when red has fallen red 67 times out of the last 100 spins. Or, to play more often, you should bet on black when red has fallen red 7 times from the last 10 spins).
So these are the conditions (or better) I was going to highlight.
In other words for eagle for combinations of 3 spins will need to solve the following system (taking into account the transition to the sequence falling out 1-8) .
First system:
There is a sequence of x1, x2,x3,x4, x5
Where and x1 and x2 and x3 and x4 and x5 belong to the set [1;8]
You must find all sequences of these numbers from 1-8
At the step of x4 we have to find moments (such sequences x1x2x3x4)at which
|x4-x3|-|x3-x2|<0, |x3-x2|-|x2-x1|<0, and |x5-x4|-|x4-x3|> or = 0 at this we need X5 to
has as few solutions as possible so that x5 (it is not known at this stage) can be of 1 or 2 -x values from the set[1;8]
The second system
there is a sequence of x1, x2,x3,x4, x5
Where and x1 and x2 and x3 and x4 and x5 belong to the set [1;8].
You must find all sequences of these numbers from 1-8
At the step of x4 we have to find moments (such sequences x1x2x3x4)at which
|x4-x3|-|x3-x2|>0, |x3-x2|-|x2-x1|>0, and |x5-x4|-|x4-x3|< or = 0 at this it is necessary that x5
Have as few solutions as possible, so that x5 (it is not known at this stage) could be of 1 or 2 -x values from the set[1;8]
Then we look, let's assume we have such conditions at step x4, i.e. the probability of x5 equal to 5 or 7 is very high, go back to step 1 and look for the numbers 5 and 7, and look for common points, let's say we assigned 110 to 7, and 101 to 5.
Based on this, we calculate the rates.
ITEM 3 Remember that in point 1 we arbitrarily assigned a series of 3 hits a number from 1 to 8. Now we make permutations, i.e. we change the assignment from 1 to 8 for the series, and repeat points 2 and 3.
Increasing the series for 4 of heads and tails combinations becomes 16, for 5 more, and so on. This is very difficult to solve with brute force. Therefore it is necessary to make a table, calculate in advance, and then when the necessary conditions appear stupidly act according to the plan.
Note that statistically even for roulette of 37 numbers if |x4-x3|-|x3-x2|>0, |x3-x2|-|x2-x1|>0, the probability that |x5-x4|-|x4-x3< or = 0 will be higher
Note that statistically even for a roulette of 37 numbers if |x4-x3|-|x3-x2|>0, |x3-x2|-|x2-x1|>0, the probability that |x5-x4|-|x4-x3|< or = 0 will be greater
Here is an example of useful properties that should be taken into account when working with series (I have attached a file with a chart)
The graph shows these properties (i.e. the statistical advantage of those conditions falling out
roulette of 37 numbers if |x4-x3|-|x3-x2|>0, |x3-x2|-|x2-x1|>0, the probability that |x5-x4|-||< or = 0 will be greater
and |x4-x3-||-|x3-x2|<0, |x3-x2|-|x2-x1|<0, the probability that |x5-x4|-|||> or = 0 will be greater )
In the file in the first column is random from 1 to 37 by an excel generator, however you spin the graph for determining the size of the increment by these conditions will crawl upwards.
here a little bit about these properties.
https://www.mql5.com/go?link=http://alu5.yjxlt.qjtv.e/go?http://eqysnk.6hfejvra.owl.e/showthread.php&t=37629&page=22 (post by henium
By the way, realising that you don't need to predict anything in a random flow (just unpredictable) has allowed me to build a trading system that can basically squeeze out any 'reasonable' return. Reasonable in the sense of not getting caught on the ass and being asked to change broker or get out of the markets altogether, as well as in the sense of deposit size.
We have to exploit the available EVIDENCE properties, which turns out to be the same in random flows and in forex ones. "Obvious" properties I for some reason only saw when I made the GSH with Eurobucks distribution. Although almost everyone knows about them. And I knew them before the LFG, but I did not immediately understood that the happiness is in them.
Nah! I've been doing my DS for four years. I got cholecystitis, gastritis and some dermatitis. I almost poisoned each other with Spider here. So please yourself. The basic ideas I have already laid out here (I mean, in this Nicholas thread). Read carefully!
I'll only repeat, that at the heart of it is the idea of increasing the lot after a loss. The inevitability of pullback is exploited. Losses are taken into account and are a system parameter. It's like nothing new, isn't it? And for calculation of the minimum necessary lot size, the size of Profit and Loss of the next trade, some target function (complex, there are different variants) is used. The most freaky variant with predetermination of the required profitability for a period. Of course, at this function the requirements to a deposit greatly increase, but if the market is not very trendy, the system mows points so that it can cause eyebrows to roll.
To Used
Looked at the file - what is it?
How about a new branch?
Many people calculate the probability from the skewed series of pros and cons.
But I see it in a different way. Calculate the probability (not even probability, but I do not know what to call it) of the series falling out through not through the increments themselves, but through modules of increments.
A series of numbers can be smaller and smaller, creating negative increments in a sequence, but the DIFFERENCE OF MODULES OF PRIRACES changes its sign much more often than the increments themselves.
In other words, probabilities of volatilities are much more profitable than probabilities of increments themselves.
I agree that these properties are present in both quotes and SB. The drunken motros formula that connects steps strangely enough works for the sizes of increments in both SB and Forex.
Not about forex yet though....
So it is possible to find it for everything, both for roulette with 37 numbers, and for eagle's game with only 2 numbers, and for everything else. Or you can have a game of roulette and transform it into roulette, or roulette into orlette.
For some reason I've tried to adapt and profit from these EXCELLENT properties just through combinatorics.I'll describe it point by point
ITEM 1.
Let's say there is an eagle. Take all variations of 3 outcomes, there will be 8 of them, i.e. 1-8.
And we turn the orlyanka sequence into a sequence of these 8 numbers.
We give to each combination a number from 1 to 8 .
ITEM 2
Further we make a difference in increments from this sequence from 1-8 ... And we see that statistically it is advantageous to bet on a roll + or 0 when two consecutive minuses have fallen out. Similarly for pluses.
Suppose we had 8 7 5 2 (combination is rare but don't pay attention it is for clarity)
Construct a sequence of incremental modulus differences.
|7-8|-|5-7|,|5-7|-|2-5|.... i.e. -1,-1,... It becomes more probable that the next number will be with + or 0. But plus can only be if the number 8 or 7 or 6 or 5 or 4 or 3 comes up.
So you need to search such combinations in which the number of such fallout will be 1 or 2, ie, the probability of falling out of 8 combinations of 2 or 3 will be very high.
ITEM 3
Then we look at the combination and make a capital allocation for this system of 2 or 3 series.
But of course such cases are rare, so we need to find as many of them as possible.
In the very first section, we assigned numbers from 1 to 8 to the series, but we did it arbitrarily. Now we go through all combinations of assignments of numbers from 1-8 to these series, and repeat the rest of the calculations. There is little logic in this, but still strangely enough it is justified by results, but mathematically I do not know how to justify it.
Point 5 . We have only done this for series of 3 eagles and tails, and there are a lot of variations.
In this way we can calculate PATTERNS for rarer outcomes for series of 3 throws, for 4... and so on. There are not enough resources to calculate all this online. That's why we need to calculate rare and probability patterns that are completely logically unrelated to each other. There will be a kind of series of patterns (not strict) on the achievement of which we enter into action.
But at the same time these unconnected patterns will work in parallel and each will have a separate piece of capital to work with.
PS: By the way, tick activity, to some extent also determines the composition of volatility, but sometimes creates interesting anomalies.
As the man writes about here
artikul :
One more example ))) The coefficients of growth of tick volumes and coefficients of price changes are compared))) Entering or exiting the market occurs two bars before the formation of the next fractal)))
artikul :
I think it's the end result that matters in the business we all do. It's the result, not whether you stick to someone else's beliefs or believe blindly in other people's theories. That's why all that matters is what you can or cannot do as a trader and what your TS is capable of. And even though DTs have a lot of stones hidden away, they are not all-powerful. The phenomenon of the market in terms of incoming information is that it is a self-organizing structure, which is isomorphic to itself in each of its fragments. And even if (as has already been said here) DTs deliver neutered quotes, there is still nothing they can do with the information, which will one day inevitably self-organize.
Here is an example from my tests. Two different brokerage companies, two terminals, the same pair, the same TF, the same TS. In one brokerage company trade at 4-mark and in another one at 5-mark. Looking. On the same bar one TS opens to sell and the other to buy. What turned out. On the 5th sign, the TS got into a corrective wave, normal worked it off, closed the profit and opened to buy. On the 4th sign, the indicator did not even move and showed a continuing trend. The whole correction wave from the 5th sign, on the 4th sign, looked like a long lower shadow of a candlestick bar. That was it. )))
Once again. Absolute values of tick volumes by themselves are no more valuable than inscriptions on the fence. But paired with the close price they form structures which must be analyzed. In this regard there is a transition from a linear quantitative analysis (price only) to a non-linear qualitative analysis (price/volume), which is worse than all financial mathematics, but incomparably more profitable. )))
In the struggle between the rake and the forehead the rake always wins. Good luck ))))
artikul :
Volumes allow you to identify non-contradictory structures (subsets) of price bars that correspond to either a reversal or a strong continuation of the trend . ))) The values of volumes by themselves are not very informative and not very useful. The effect is achieved when the closing price in combination with the volume forms an anomaly. )))
But it's not that simple. You should not rush headlong into the ticks.
We need to play on the probabilities of occurrence of signs of incremental modulus difference, which has quite working properties, but this probability is also characterized to some extent by tick density, or real tick volumes, which correlate with just the number of ticks occurrence. And sometimes the juxtaposition of volatility analysis through tick density and through increment modulus sizes can be useful.
Beautiful! A whole article is piled on top of it!
But interesting to read and makes you think again about some things. Thanks.
Wow! That's a bit complicated. Maybe something simpler, for example eagle, three throws, we have 8 possible variants, we wait for the variant OOO or RRR and then we know that in the next three throws the same 1/8 series is repeated.
jelizavettka:
Просто попробуйте для начала торгануть график серий.
aaaaah that's what you mean, thank you, that's really interesting to think about))).