Machine learning in trading: theory, models, practice and algo-trading - page 3050

 
Aleksey Vyazmikin obtained randomly, i.e. it does not describe a long-term pattern.

Thus, we obtain a set of rules+predictors, which can reveal the regularity on our time series.

Will someone try to do this in R or Python?

I even posted screenshots........

I practically wrote a lecture on this topic on the forum (in posts)...

There is a pattern, and it's so strong that it's unbelievable.

Not even a regularity, but a direct dependence.

I even published the formula (equation) here in the branch.

 
Renat Akhtyamov #:

I even posted screenshots........

I practically wrote a lecture on this topic on the forum (in posts)...

There is a pattern, and such a pattern as to make a mother of a lot of trouble

Not even a regularity, but a direct dependence.

I even published the formula in this thread.

I hope you kept the post number and kindly post a link to the pattern now so we can finally start earning and get distracted on the forum just because of boredom.

 
Ivan Butko #:

I hope you kept the post number and kindly post the link to the patterns now so we can finally start earning and distracting ourselves on the forum just because of boredom.

I found one post, but apparently there was something else before it.

Where have you seen investing without drawdowns, drains or risk of draining? -MQL4 and MetaTrader 4 - MQL5

Эксперимент - Где вы видели инвестирование без просадок, сливов или риска слить?
Эксперимент - Где вы видели инвестирование без просадок, сливов или риска слить?
  • 2021.08.11
  • www.mql5.com
Итог должен получиться такой - работа на весь депозит и чем выше плечо. Никто не отважится вложиться на убыточный сигнал только под уважаемого форумянина. Да никто в здравом уме в такой сигнал денег не засунет
 
Renat Akhtyamov #:

I found one post, but there must have been something else before it.

Where have you seen investing without drawdowns, drains or risk of losing? -MQL4 and MetaTrader 4 - MQL5

Great, thank you

 
Many years ago, when I just started trading, I had a million ideas in my head to create a TS.
I wish I could find a programmer, I thought, I have so many ideas, but a programmer has a good skill, but no ideas... I would invent and he would code....

What an idiot I was back then....

Now I realise that a programmer has both skill and experience and quality ideas based on real experience.

And the first one just has dream ideas based on fantasies and isolated observations.
 
Aleksey Vyazmikin obtained randomly, i.e. it does not describe a long-term pattern.

Thus, we obtain a set of rules+predictors, which can reveal the regularity on our time series.

Will someone try to do this in R or Python?

Looks like I have to do it all myself again...

 
Aleksey Vyazmikin #:

I guess I'll have to do it all myself again.

I don't get it. I don't have time yet. What kind of graphs to generate? Can we filter errors not only from the symbol on which we trained, but also from other (say, correlated) symbols? Maybe it makes sense. Ah, I've already done that. No sense :)
 
Maxim Dmitrievsky #:
I don't get it. I don't have time for this. What kind of graphs to generate? Can we filter errors not only from the symbol on which we trained, but also from other (say, correlated) ones? Maybe it makes sense. Ah, I've already done that. No sense :)

I understood it as a search for identical patterns among different instruments. And then test them on 100 rows of SB. If the found rules will work on the SB rows, then reject them.

 
Maxim Dmitrievsky #:
I don't get it. I don't have time for this. What kind of graphs to generate? Can we filter errors not only from the symbol on which we trained, but also from other (say, correlated) symbols? Maybe it makes sense. Ah, I've already done that. No sense :)

Not necessarily correlated. Why wouldn't it make sense? I took quant segments on EURUSD and tested them on GBPUSD - about 25% showed probability bias on both pairs. What happens if you take another pair - I don't know, yet. Of course, there is a small problem - I am working with the basic signal of the strategy, and whether it should be changed on other pairs or not for this study is not clear yet. Obviously, the nature of different instruments is different. Perhaps the instruments need to be clustered at the beginning, or otherwise grouped. In general, the question of setting up the experiment is open.

Generate charts, apparently, taking into account the average descriptive statistics of a group of trading instruments. I.e. something similar in skeleton, but with random fat.

 
Valeriy Yastremskiy #:

I understood it as a search for the same patterns, patterns among different instruments. And then check them on 100 SB rows. If the found rules will work on the SB rows, then reject them.

All right, thanks for understanding.