Looking for patterns - page 97

[Tretyakov Rostyslav]

It's kind of quiet :(

[khorosh]

MakarFX:
It's kind of quiet :(

"All is frozen again until dawn..."

[Evgeniy Chumakov]

MakarFX:
It seems to be quiet :(

Apparently they found a pattern ;)

Here is a study on how many extrema are updated in a time interval.

The essence - if the current maximum or minimum for the period of 20 minutes (that is what I have chosen) passes through the previous maximum and minimum for the same period (by a shift of 1 minute), it means that the extrema have been updated. We count how many updates there have been in the time interval of 20 minutes and save the values.

The histogram on the left shows the number of updates and frequency, and the interval in minutes since the last update of extrema is on the right.

[Evgeniy Chumakov]

Or here's one on the same topic, only here we take a zigzag with the Depth parameter 24. The question is how many extrema fall in the time interval of 24 minutes.

We can see that the time interval of 24 minutes often contains 0 to 1 extremum, and less frequently 2 and 3 extremums.

It may be more interesting for us to identify the situation when there is no update of the zigzag in the next time interval. But I suspect, judging from the histogram, that it is 50/50.

[Tretyakov Rostyslav]

Evgeniy Chumakov:

Or here's one on the same topic, only here we take a zigzag with the Depth parameter 24. The question is how many extrema fall in the time interval of 24 minutes.

We can see that the time interval of 24 minutes often contains 0 to 1 extremum, and less frequently 2 and 3 extremums.

It may be more interesting for us to identify the situation when there is no update of the zigzag in the next time interval. But I suspect, judging by the histogram, that it is 50/50.

I don't quite get it, but you can do the same research on a larger M5 or M15 timeframe.

The M1 is very noisy, imho.

[khorosh]

MakarFX:

I don't quite understand, but you can do the same research on a larger M5 or M15 timeframe.

M1 is very noisy, imho.

There is an indicator close to the question at hand.

[Tretyakov Rostyslav]

khorosh:

There is an indicator close to the question at hand.

You are talking about this indicator here version v3.

[Evgeniy Chumakov]

MakarFX:

I don't quite get it, but you can do the same research on a larger M5 or M15 timeframe.

M1 is very noisy, imho.

OK, here's the zigzag for H1 Depth = 8 Hours (Sort of a session).

Who would tell me what kind of distribution to use to calculate the probabilities? Or better still, do the research yourself.

[Renat Akhtyamov]

Uladzimir Izerski:

What is the point of watching currencies.

If a currency is rising we do not sell it. A clear example at the moment 17:33,02.03.2020 local time and compare it with yesterday's picture.

The waves carry important information about the future price. The pictures below are taken a few minutes later. They are essentially military radars.

Oooh, this is where it gets interesting.

I recognise the algorithm.

at least the sum is zero.

[Martingeil]

Has anyone tried using Volume, is it worth using it in calculations?

       double max_hig = MathMax(MathMax(high[1],high[2]),high[3]);
       double min_low = MathMin(MathMin(low[1],low[2]),low[3]);
       double ran_hl  = max_hig - min_low;
       
       long   max_vol = MathMax(MathMax(iVolume( _Symbol, 0, 1),iVolume( _Symbol, 0, 2)),iVolume( _Symbol, 0, 3));
       long   min_vol = MathMin(MathMin(iVolume( _Symbol, 0, 1),iVolume( _Symbol, 0, 2)),iVolume( _Symbol, 0, 3));
       long   ran_vl  = max_vol - min_vol;