Discussion of article "Developing stock indicators featuring volume control through the example of the delta indicator" - page 2
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
I started the indicator with default settings, but the retrospective calculation was made only for today's date - TF minutes - why is it so?
Before asking any questions, please read the article carefully:
Point 4: "Set the moment when ticks of generated bars start loading". Getting tick history can be a rather time-consuming operation. Therefore, it is necessary to allow the user to specify the start date of its downloading. The indicator provides the inpHistoryDate parameter for this purpose. If the value is zero, the history is loaded from the beginning of the current day. In this prototype of the SetFrom(datetime) method the time is passed in seconds. As mentioned above, the calculation of the formed bars of the indicator is performed first.
I get it, it's a common situation - what's the reason for that?
About the formed bar - no, I'm just interested in how the EA determines that a new tick of a potentially new bar has arrived, although the bar is not formed yet.
Reread my post #7 again.
In order to determine that a new tick of a bar that has not been reflected yet has arrived, you need to understand the algorithm used in my indicator. It is described in detail. Namely: get the tick history and compare the time of each tick with the opening time of the last bar.
Before asking any questions, please read the article carefully:
It's not a zero value! I took it, started it, and I don't see zeros - so there is no association, went on to the OK button. Just now I noticed that the date is ancient....
Re-read my post #7 again.
To determine that a new tick of a bar not yet reflected has arrived, you need to understand the algorithm used in my indicator. It is described in detail. Namely: get the tick history and compare the time of each tick with the opening time of the last bar.
Thank you for your reply.
It's not zero! Here, I took it, ran it, and I don't see zeros - so there is no association, went on to the button ok. Just now I noticed that the date is so old....
1970.01.01.01 00:00:00 - this date = 0 for datetime type.
1970.01.01.01 00:00:00 - this date = 0 for datetime type.
Of course it is, from the point of view of the code, and I evaluated by the fact what it shows. It doesn't matter, we've sorted it out and it's good.
Further development of the indicator is seen in the construction of muwings for deltas and finding the delta between them, and deltas for averaging should be taken not by bars, but by values.
Of course it is, from the point of view of the code, and I evaluated by the fact that it shows. It doesn't matter, you've figured it out and that's fine.
It is seen not from the point of view of my code, but from the point of view of documentation.
Further development of the indicator is seen in the construction of mooings for deltas and finding delta between them, and deltas for averaging should be taken not by bars, but by values.
You and I see the code development differently. You can analyse the market much more thoroughly on the basis of tick information than with simple swipes. For example:
And you talk about averaging... There is pure, detailed information. You should not average it. IMHO, of course.
This is not seen from the perspective of my code, but from the documentation.
Of course, your code is written according to the documentation, how else!?!!?
You and I see code development differently. On the basis of tick information you can analyse the market in much more detail than with simple mashes. For example:
And you talk about averaging... There is pure, detailed information. You should not average it. IMHO, of course.
Are you going to analyse it all by eye? I am talking about instruments, but now an indicator is just a good separator of information into two groups, nothing more.
Of course your code is written according to the documents, how else!?!!?
Are you going to analyse it all by eye? I'm talking about tools, now an indicator is just a good separator of information into two groups, nothing more.
And averaging values gives you a tool that can be applied immediately?
Of course, if you want to see a "delta surge", you need to see some previous history and have an average value of this history to know how strong the surge was. But you don't have to average the delta to do that. You can simply visually determine the level after which a spike is considered strong. For RTS, for example, this value is >= 1000.
Does averaging values immediately give you a tool to apply?
Of course, if you want to see a "delta spike", you need to see some previous history and average that history to know how strong the spike was. But you don't have to average the delta to do that. You can simply visually determine the level after which a spike is considered strong. For RTS, for example, this value is >= 1000.
For me, volumes in such a representation are not yet a studied phenomenon - I am just sharing my thoughts out loud. My proposed variant should show a trend, and if it will be an alternative variant of TA and other indicators based on OHLC, why not use it as an independent tool?
Visual (here I mean in the mind) is good, but it is necessary to translate the whole thing into digital form to test hypotheses and continue generating new ones.