From theory to practice - page 1434

 
khorosh:

He's talking like a dissertation.)))

But these words are of little use, because he did not take into account the redistribution of Gus Lapluson: money in the stock market is redistributed from the smart and fools to the fast and wise. I.e. it's either the sucker to the stock exchange or the gut that makes money. ))

 
khorosh:

That's quite a thesis.))

:))) Well, that doesn't make it any easier.

SB is as difficult to fight as it was and will continue to be.

And on ticks - still an unexplored theoretical area. But maybe it will dawn on someone... Maybe neural networks will crack the ticks... I don't know.

 
Alexander_K:


The proof of randomness of the process is that the sum of a large number of independent random numbers belongs to a normal Gaussian distribution.


Yoaklmn, "physicist" ....

Why you?

 
Alexander_K:

OK,

........

So, the mechanism of formation of tick increments is as follows:

a) at time t1 the broker has a set of prices for a currency pair (I don't know, correct me here, but I think it is called " Depth of Market") and this set is a Poisson distribution.

b) from this set the broker randomly selects one price - a tick is generated.

......

Thanks for your attention.

stop, not like this

if some price is selected from this set, then the price will pass through some other prices from the same set as well

you needn't read any further, because the beginning of the logic of the randomness rationale has already been buried under the ice.

Moreover, there will be not a tick but a series of ticks

In order to make your hypothesis even slightly similar to practice, it will certainly be necessary to cut out (thin out) this series of ticks

---

I've added another line of myth busting to my piggy bank

 

It seems to me that there is an agreement among brokers to have information about all open orders and there is a certain official channel for the price (set by the banks of the country) and everything inside the channel should be taken from traders.

The channel for EURUSD, for example, is based on data from the European bank and the Fed.

 
Evgeniy Chumakov:

Somehow it seems to me that there is an agreement among brokers to have info on all open orders and there is a certain official channel for prices (set by the country's banks) and everything inside the channel should be withdrawn from traders.

The channel for EURUSD, for example, if you build it according to the data of the European bank and the Fed.

there is one, it is the CME.

 
Дмитрий:

Fucking "physicist" ....

Why are you doing this?

What's your question, hillbilly?

 
EgorKim:

I think the problem is that the market is chaos and the market has no memory.

Guys...

Well, I can understand when some distant from the practice "academics", in an attempt to gain a reputation in the circles of their peers, re-citing everyone and confusing to the point of absurdity, with foam at their mouths prove that the market is efficient, but their job is to be scientific, obscure, mysterious, they are on the payroll. But what about you?

What the fuck "efficiency" if all who make transactions on the real market do so with hope and trepidation, having invested their whole soul and mind. And all participants (both people and algorithms) look at the history, analyze approximately the same events and draw similar conclusions, taking into account of course the gaming, competitive nature of the market, understanding that too obvious trends are seen by many and they can be a trap. In general,the market has a memory for sure, it is not even discussed, but it is not the same as the memory of a stone in space that keeps its speed through time, it is the memory of the trajectory of a football during a game of football in the big leagues, or the memory of chess pieces during a game of grosmeister. There is no point in dissecting such processes with DSP orlinear regression, because the price itself is as informative as the pure trajectory of a football ball, without taking into account the location of players on the field and their skills, etc., everything is much more complex and interesting.

 
Грааль:

Guys...

Well, I can understand it, when some far from practice "academics", in an attempt to gain a reputation in the circles of their own kind, by quoting everyone and confusing them to the point of absurdity, are frothing at the mouth and proving that the market is efficient, but that is their job, to be scientific, obscure, mysterious, they are on the payroll. But what about you?

What the fuck "efficiency" if all who make transactions on the real market do so with hope and trepidation, having invested their whole soul and mind. And all participants (both people and algorithms) look at the history, analyze roughly the same events and draw similar conclusions, taking into account of course the gaming, competitive nature of the market, understanding that too obvious trends are seen by many and they can be a trap. In general, the market has a memory for sure, it is not even discussed, but it is not like a rock in space that keeps its speed through time, it is like a memory of the trajectory of a football during a game of Major League Soccer or the memory of chess pieces during a game of grandmasters. There is no point in dissecting such processes with DSP or linear regression, because the price itself is as informative as the pure trajectory of a football ball, without taking into account the location of players on the field and their skills, etc., everything is much more complex and interesting.

The paradox of the market is precisely that the regular, dependent tick increments, when evenly thinned, give a random process forOPEN/CLOSE M1, M5, ...

And while it's difficult to capitalize on ticks with formalized methods - you should actually create a mathematical apparatus of non-Markovian processes for that, or use a neural network to reveal the tick series, then there is already some mathematics for the random process - the probability theory and DSP.

Koldun's indicator, for example, easily copes with SB, but with real BP onOPEN M1 - much harder.

Otherwise Zhenya Chumakov who used it would have become a millionaire long ago. Still, trends (deterministic movements with memory) are tearing him up too.

Or am I wrong? If Zhenya is reading these lines, can he demonstrate the tests?

 
Alexander_K:

The paradox of the market consists exactly in the fact that regular, mutually dependent tick increments, when thinning them out, give a random process for OPEN/CLOSE M1, M5, .

And while it's difficult to capitalize on ticks with formalized methods - you should actually create a mathematical apparatus of non-Markovian processes for that, or use a neural network to reveal the tick series, then there is already some mathematics for the random process - the probability theory and DSP.

Koldun's indicator, for example, easily copes with SB, but with real BP onOPEN M1 - much harder.

Otherwise Zhenya Chumakov who used it would have become a millionaire long ago. Still, trends (deterministic movements with memory) are tearing him up too.

Or am I wrong? If Zhenya is reading these lines, can he demonstrate the tests?

Sash, what tests?

I have applied the trades to the sorcerer's indicator and there is a 50/50 win/loss.

My aim is to get the trades to open and close.

We should use the indicator on the chart as I showed it yesterday and see what is there.

The prerequisite must be that there are no sell trades below the buy signal and vice versa, i.e. no buy signal above the sell signal.