ZigZags Shepherds - page 5
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
Thanks, I'm still on 4, I need to slowly move to 5, my broker will be using 4 by the end of the year, and 5 has already been introduced.
my broker is finishing using 4 by the end of the year, and has already introduced 5.
there is no way that a broker would give up mt4.
show the news about it on the broker's official website.
There is no way that a broker would give up mt4.
Show the news on the broker's official website.
That's what I was told by the customer service, it seems it's forbidden to advertise a broker here, especially one in the Polish market.
Why is there research here? It is enough to remember that some brokerage companies give 4 spreads, and some give 5. Ticks are fully determined by each brokerage company for each account type and sometimes individually for each account. Roughly speaking, the area of fluctuations up to 2 spreads is fully managed by brokerage companies that decide how to divide rate movements and may divide them 10-fold or packets with more or less ticks, for example from 80 thousand to 1500 thousand per week.
And the exchange rate movement analysis you cited with an exponential pattern is for rate differences, not for ticks.
Yes thanks, read already in this thread https://www.mql5.com/ru/forum/221552/page205#comment_6733812 ))
This research wasn't done by me, Northwind.
Probably quite tired by now, can you clarify in what sense for the rate difference?
Yes, thanks, already read in this thread https://www.mql5.com/ru/forum/221552/page205#comment_6733812 ))
That research wasn't done by me, Northwind.
Probably quite tired by now, can you clarify in what sense for the course difference?
In your two lines:
"Where+-1 accounts for 99.4% of all ticks and +-2 accounts for about 0.55% more.
It turns out to be an inconsistency. I.e. 1H intervals for 1p would be more than 50%."
By a "tick" (+-1 point, the value of a point is at the discretion of the brokerage company) we mean the increase of the exchange rate during the period of time between two successive exchange rates. The size of the interval:
- is orders of magnitude shorter than the characteristic time of formation of the "move" step or the threshold value of the rate increment in the second line, seconds and milliseconds compared to minutes and hours for the intervals of the second line;
- is not determined by the threshold value (not the course stroke), but by the selection and filtering algorithms applied in the DC.
In addition, there is a rigid relationship between increments and rates, the sum of tick increments for any period equals the rate increment for that period. Not only the value, but even a sign of increments can move at will of brokerage companies. But in case of more than 2 spreads the DC will create an arbitrage situation and will be punished by the trading systems that calculate them. Materially.
The thresholds of the second line are chosen by the client, not the DC, and their values are usually more than several spreads.
From the standpoint of mathematics, the mentioned rigid relationship can be expressed by the analogy between the values of a table-defined function and the integral sum over the table, if we consider the table row number as an integration variable. Recalling the relation between the definite integral and the first form of the subintegral function, we can look at the relation of the tick increments to the rate as well as the relation of the increments of a function to its derivative directly in the definition of the derivative as the limit of the ratio of the increments of the function and the argument. Only you can't go to the limit. And difference substitutions for the derivative do not work well because of chattering. If we exclude (decrease) the dithering of tick increments, for example by averaging the average rate 0.5*(Bid+Ask) over dozens of DC sources we can get very close to it.
One more analogy, which is purely my own. Looking at a thermometer outside the window we say: "Wow, it's ten degrees colder this morning than it was last night". These are the differences in the course the Northern wind (the author of the graphs) operates in the second line. But if we come to a shop, ask for an outdoor thermometer with -50+50 school and we will be given a choice of five pieces from the same box, the difference in their readings, even two or three degrees is analogous to tick differences. This is their area, take a look at the now common quote comparison tables from different DCs.
In your two lines:
"where+-1 accounts for 99.4% of all ticks, and +-2 accounts for about 0.55% more.
This results in an inconsistency. I.e. 1H intervals for 1p. would be more than 50%."
By a "tick" (+-1 point, the value of a point is at the discretion of the brokerage company) we mean the increase of the exchange rate during the period of time between two successive exchange rates. The size of the interval:
- is orders of magnitude shorter than the characteristic time of formation of the "move" step or the threshold value of the rate increment in the second line, seconds and milliseconds compared to minutes and hours for the intervals of the second line;
- is not determined by the threshold value (not the course stroke), but by the selection and filtering algorithms applied in the DC.
In addition, there is a rigid relationship between increments and rates, the sum of tick increments for any period equals the rate increment for that period. Not only the value, but even a sign of increments can bounce at will of brokerage companies. But in case of more than 2 spreads the DC will create an arbitrage situation and will be punished by the trading systems that calculate them. Materially.
The thresholds of the second line are chosen by the client, not the DC, and their values are usually larger than several spreads.
In terms of mathematics, the mentioned rigid relationship can be expressed by the analogy between the values of a table-defined function and the integral sum over the table, if we consider the table row number as an integration variable. Recalling the relation between the definite integral and the first form of the subintegral function, we can look at the relation of the tick increments to the rate as well as the relation of the increments of a function to its derivative directly in the definition of the derivative as the limit of the ratio of the increments of the function and the argument. Only you can't go to the limit. And difference substitutions for the derivative do not work well because of chattering. If we exclude (decrease) the dithering of tick increments, for example by averaging the average rate 0.5*(Bid+Ask) over dozens of DC sources we can get very close to it.
One more analogy, which is purely my own. Looking at a thermometer outside the window we say: "Wow, it's ten degrees colder this morning than it was last night". These are the differences in the course the Northern wind (the author of the graphs) operates in the second line. But if we come to a shop, ask for an outdoor thermometer with -50+50 school and we will be given a choice of five pieces from the same box, the difference in their readings, even two or three degrees is analogous to tick differences. This is their area, take a look at the now widespread tables comparing quotes from different DCs.
I agree with you, I was not correct initially when comparing tick and step, I did it consciously, with the probability that step in such scale will be equal to tick, in general, everything is clear. You have explained everything very well, and I thank you very much, the only thing that bothers me is your phrase"to combine packs with ticks more or less often, for example from 80 000 to 1500 000 per week".
If the small horizon chattering is within 2 spreads, why does it give such a big difference for a long period? I.e. the error in time is smoothed with the period (in bars), but it increases in ticks?
You write that by averaging the average rate over dozens of brokerage companies we can decrease the reliability. What is better, to average spread from different sources or to work with ticks from different sources?
I agree with you, I initially made an incorrect comparison of tick and pitch, I did it deliberately with the probability that the pitch in such scale will be equal to tick, in general, everything is clear. You have explained everything very well, and for that I am very grateful, the only thing that bothers me is your phrase"to combine packs with ticks more or less often, for example from 80 000 to 1500 000 per week".
If the small horizon chattering is within 2 spreads, why does it give such a big difference for a long period? I.e. the error in time is smoothed with the period (in bars), and in ticks it increases?
You write that by averaging the average rate over dozens of brokerage companies we can decrease the reliability. Which is better, to average spread from different sources or to work with ticks from different sources?
Estimates of the number of tick packs per week are taken from the figure from https://www.mql5.com/ru/forum/221552/page18#comment_6167098, based on the fact that one pack comes with no more than one tick per pair, that's the way this tick collection works. I don't see any correlation with a range of 2 spreads. What is "timing error" I don't know. I haven't written about credibility approximation and won't even say what it is.
If I knew "which is better"... Why calculate average spreads from multiple sources, I don't know. Tics from different sources are good at least for identifying arbitrage situations.
Estimates of the number of tick packets for the week are taken from the figure from https://www.mql5.com/ru/forum/221552/page18#comment_6167098, on the basis that in one packet comes not more than one tick per pair, this is how this tick collection works. I don't see any correlation with a range of 2 spreads. What is "timing error" I don't know. I haven't written about credibility approximation and won't even say what it is.
If I knew "which is better"... Why calculate average spreads from multiple sources, I don't know. Ticks from different sources are good at least for identifying arbitrage situations.
the "error in time" grows - I meant that different brokers, which follows from your table, also have a strong scatter in volumes, i.e. if initially there is a difference in the process of quoting in ticks, then over time this difference will only grow, in bars on the contrary, the more time passes, the less this difference will be. Approximation of reliability meant that if we receive the final product at the worst prices also filtered, averaged, then if we take quotes from different brokers, we may see at least an approximate picture of lag, because some brokers may broadcast the filtered flow of their liquidity provider, it is only a thought.
It turns out that when we open a position with a broker, in most cases we are immediately in the resonance, so maybe there is a higher probability for opening a position against the movement. Who has an opinion?
Here
http://www.gurutrade.ru/forex-quotes/
you can see quotes from different brokers.
The "time error" increases - I meant that different brokers, which follows from your table too, have a strong variation in volumes, i.e. if initially there is a difference in the quotation process in ticks, then with time this difference will only increase, in bars on the contrary, the more time passes, the less this difference will be. Approximation of reliability meant that if we receive the final product at the worst prices also filtered, averaged, then if we take quotes from different brokers, we may see at least an approximate picture of lag, because some brokers may broadcast the filtered flow of their liquidity provider, it is only a thought.
It turns out that when we open a position with a broker, in most cases we find ourselves immediately in the resonance, so maybe there is a higher probability for opening a position against the movement. Who has an opinion?
We can see both a lag pattern and an advance pattern. Only:
- these patterns are instantaneous, on the next poll of incoming ticks the leading DC will become the lagging one;
- The scale of the differences in these patterns is small, otherwise arbitrage situations will arise.
What is the difference in the number of ticks in one and the same time I understand. And what is a difference in the number of bars, for example, of fifteen-minute bars?