Machine learning in trading: theory, models, practice and algo-trading - page 3596

 
Aleksey Nikolayev vertical lines) is interesting.

On gold, for example, such peaks are much weaker.

Perhaps it is time to add round price levels to round time levels)

On different DCs, look at the beginning. This is either the first hour of the day or news.

 
Aleksey Nikolayev vertical lines) is interesting.

On gold, for example, such peaks are much weaker.

Perhaps it is time to add round price levels to round time levels)

People's daily routines are timed to round numbers. The office opens at 8:00, the boss has a meeting at 9:00, and so on. And all technological-trading processes inherit this.
Time zone differences are massively multiples of 1 hour, regular events of stock exchanges and banks are also multiples of hours.

 
Aleksey Vyazmikin #:

On different DCs, look at the beginning.

Makes sense. Doesn't explain the difference with gold though.

Aleksey Vyazmikin #:
news.
This is the average behaviour for all days since 01.01.24. News is obviously not released every hour of every day.
Aleksey Vyazmikin #:
first hour of the day

There are understandable large peaks in the first hours. The question about hourly small ones.

Maxim Kuznetsov #:
people's daily routine is timed to round numbers. The office opens at 8:00, the boss has a meeting at 9:00, and so on.

And so every hour during the day and for some reason in euros, and gold seems to be traded by slackers).

 
I want to manage a hedge fund using MO algorithms and I want to play diablo 4 on my console with a friend. I want to write an article to be famous and don't want to reveal different features that lead to success. I want to hit the moderators, but I don't want to be banned. Constant problem of choice :)
 
Aleksey Nikolayev vertical lines) is interesting.

On gold, for example, such peaks are much weaker.

Perhaps it is time to add round price levels to round time levels)

The further from zero the average spread is, the weaker are its average spikes.

ZY I have launched a spread indicator from KB showing the average (time-weighted average) spread at each minute for the last 250 days. It did not show anything similar in terms of spikes on EURUSD.
 
fxsaber #:

The further from zero the average spread is, the weaker its average spikes are visible.

ZY I ran a spread indicator from KB showing the average (time-weighted average) spread at each minute for the last 250 days. It did not show anything similar on EURUSD spikes.

It seems to be the peculiarities of a particular brokerage centre. I didn't see such spikes on metaquotes. When I have time, I will check it on a couple of other brokerage centres.

 
Aleksey Nikolayev #:
And so it is every hour during the day and for some reason in euros, while gold seems to be traded by slackers).

Gold is expensive and the turnover there is less than in EURUSD, so the peaks are small. But they are there. By the way, it is necessary to count their prices and only then compare them, otherwise "physics fucking it" - comparison of disproportionate values

The reason and consequence of this is from the dollar, it is the dollar's collision at the boundaries of timeframes. Purely practical - never trade on the opening of M30 H1, you can say tradition. "It is better to miss a tick or two than to catch a requote or slippage...". :-)

 
Maxim Kuznetsov #:

Gold is expensive and the turnover there is less than in EURUSD, so the peaks are small. But they are there. By the way, it is necessary to count their prices and only then compare them, otherwise "physics fucking it" - comparison of disproportionate values

The reason and consequence of this is from the dollar, it is the dollar's collision at the boundaries of timeframes. Purely practical - never trade on the opening of M30 H1, you can say tradition. "It is better to miss a tick or two than to catch a requote or slippage...". :-)

I see
 

> peculiarities of a particular DC. I have not seen such surges at methaquots

Methaquots don't trade. Figuratively, they integrate existing flows. they have very clean and very artificial quotes

 
Aleksey Nikolayev #:

It seems to be the peculiarities of a particular DC. I didn't see such surges in methaquot. If I have time, I will check a couple of other DCs.

I was not lazy to check one more brokerage house (real account, as for the initial pictures). The third variant turned out to be: if we use the arithmetic mean as the average, there are peaks, but if the mean is used as the median, they disappear).

In general, it is clear that the peculiarities of DC. So the question is cancelled (or rather changed to a question not asked here on the forum).