From theory to practice - page 10

 
Alexander Sevastyanov:

Generally speaking, futures and options are directives (derivatives) on the prices of currency pairs formed by forex transactions. For example, a bank, fulfilling a client's order, carries out a currency exchange transaction on the forex market. In this case the bank is not a speculator and does not look back on futures or options. He simply executes the client's order. Another example: the regulator (Central Bank) carries out currency intervention, it also does not look at futures/options.

IMHO for analysis you should take the tick source of a regulated forex ECN/STP broker, preferably a real server.


I disagree, our regulator comes out with currency interventions just not on forex, but on MICEX

 
Alexander_K:

Yes, that would be a REAL solution to the problem of traders communicating from different DCs.

Thank you, Nikolay!


Looking for where to get CME data cheaply and serially

 
Alexander_K:
Maybe it is easier to work with a single data provider?

For development, of course, it is better to take one supplier, and develop for it, and then finalize it with different suppliers.

Alexander, if it's convenient for you, it's no problem to work with your supplier, the main thing is to standardize so that there are no misunderstandings.

Your supplier has a tick history?

ZZZI If so, the name of the supplier, the number of the account, the invest password in the studio.

 
Nikolay Demko:

For development, of course, it is better to take one supplier, and develop for it, and then finalize it with different suppliers.

Alexander, if it's convenient for you, it's no problem to work with your supplier, the main thing is to standardize so that there are no misunderstandings.

Your supplier has a tick history?

ZZZI If so, the name of the supplier, the account number, the invest password in the studio.

I don't have it. I'm struggling with this and collecting data history myself.

Once again - archive data is VERY important. I don't even know how to emphasize its importance. Let me put it this way - VERY IMPORTANT. They are used to calculate the average historical variance and nonparametric skew in the model.

Without these averages, all calculations for non-Markovian models are meaningless.

 

As this is a general discussion, traders who are interested in this topic should have blocks in their software arsenal to calculate:

1. moving average (MA)

2. weighted moving average (WMA)

3. moving average

4. weighted dispersion

5. dispersion

6. nonparametric skew

All this should work with a sample size of at least 50,000 ticks.

Even if you have your own view of the market and do not want to understand my developments, I assure you - these tools are very useful, you must only know how to apply them. And in MQL, I don't think there is a moving average and therefore there is no nonparametric skew. How the algo-traders work without it - I don't know. :)))

 
Alexander_K:

As this is a general discussion, traders who are interested in this topic should have blocks in their software arsenal to calculate:

1. moving average (MA)

2. weighted moving average (WMA)

3. moving average

4. weighted dispersion

5. dispersion

6. nonparametric skew

All this should work with a sample size of at least 50,000 ticks.

Even if you have your own view of the market and do not want to understand my developments, I assure you - these tools are very useful, you must only know how to apply them. And in MQL, I don't think there is a moving average and therefore there is no nonparametric skew. How the algo-traders work without it - I don't know. :)))


It exists, I assure you :-)

Or rather, we have an ordinary one from which it is easy to create a sliding one.

Alexander, could you tell me more about WMA? I don't understand about the period.

 
Dennis Kirichenko:

There is one, I assure you :-)

Or rather, there is an ordinary one, from which you can easily create a sliding one.

Alexander, can you tell me more about WMA? I don't understand about the period.

And you know exactly how w weight is calculated?

So, we need a moving weighted average as in VisSim, which works with a window of data, or in other words, a sample volume.

Seehttps://www.mql5.com/ru/forum/220237/page2 for how to calculate the optimal sample size.

If we had only one data provider, I would say for EURJPY we should take the sample size 12.800, but in other words, you should calculate it for your brokerage company. And it would be great if these values are the same.

Что определяет Ваш выбор таймфрейма (ТФ) для торговли? И далее разговор о математическом обосновании выбора таймфрейма.
Что определяет Ваш выбор таймфрейма (ТФ) для торговли? И далее разговор о математическом обосновании выбора таймфрейма.
  • 2017.11.23
  • www.mql5.com
Собственная интуиция, Устоявшееся мнение сообщества, Расчет вероятности профита для разных ТФ, Привычка к данному ТФ (набил руку), Рекомендации ДЦ...
 

By the way, having re-read some threads on the forum, I understood - why the distrust of my posts and topics.

People who thought they were great mathematicians and physicists suddenly showed the most disgraceful results in practice and, in the end, everyone just got fed up. Quite instructive stories!

I see, I see... And now there's a mistrust of all "physicists" in a row... Too bad! Questioned the ability of people to achieve results by mathematical means. Well, well - this has strengthened my opinion that it is necessary to prove the opposite and to defend the honour of physicists. However, I will publish my posts less often and all of them will be on the point.

Good luck to all of you!

Regards,

Alexander_K

 
Alexander_K:

By the way, having re-read some threads on the forum, I understood - why the distrust of my posts and topics.

People who thought they were great mathematicians and physicists suddenly showed the most disgraceful results in practice and, in the end, everyone just got fed up. Quite instructive stories!

I see, I see... And now there's a mistrust of all "physicists" in a row... Too bad! Questioned the ability of people to achieve results by mathematical means. Well, well - this has strengthened my opinion that it is necessary to prove the opposite and to defend the honour of physicists. However, I will publish my posts less often and all of them will be on the point.

Good luck to all of you!

Regards,

Alexander_K


really a physicist?

 
Mickey Moose:

really a physicist?


:))) Doesn't he look like one?