Stable MTS - page 22

New comment
 
Сергей:
70/30 may change tomorrow, some MM must be. And a test on a spread of 2 on a five-digit is not informative at all. Take the five-month spread at 20. If it holds, it's a bid. I have one version of my EA showing 17 million net profits on a test with spread 2, and it pours at 20.
Why do I need a five-minute, with a spread of 20, if all the logic of the owl does not depend on ticks, and all the events take place at the opening of M1 candle. And where I trade on spread 1-3 five-digits, plus a commission of $ 8 per lot dollar? The bottom line is 11 on your spread. You have 22 initially, that is twice as much !!!
 

By the way an interesting topic, many projects with the stupid already abandoned, and yesterday collected to unity ... in any case, the best test ))) Well, time will tell.

 
Vladimir Zubov:
Why do I need a five-minute, with a spread of 20, if all the logic of the owl does not depend on ticks, and all events take place at the opening of the M1 candle. And where I trade the spread is 1-3 five-digit, plus the commission of $ 8 per lot dollar? The bottom line is 11 on your spread. You have 22 initially, that is twice as much!
Sorry, but I can't believe you've been trading it for so many years. Otherwise you would not ask such things. The slippage and widening of spreads without warning have not been canceled, especially on the real market. You may be able to do that on demo. But on the real account, especially if the dealer starts to withdraw you to the interbank ..... You may be able to ask questions or to ask questions on the real market. My personal opinion, I will not waste my time arguing. If you want to discuss it, please contact me in person.
 
Yuriy Asaulenko:

Try calculating the autocorrelation. MathLab does it in a few seconds. You'll immediately see why.

Maybe there's a better covariance, but I haven't done it.

It all depends on the number of autocorrelation points and shift range. It's probably true it will take time, but it's not like you have to count it all the time. Autocorrelation is the covariance of a time series with itself with a shifted time lag.
 
Vladimir Zubov:
But why is everyone clinging to the tests, to the expected payoff? If I always do the first test 0.01, I think it's ok around 1. What should be 100?)
And how else to check what the algorithm is capable of? There is a set of indicators that characterize the past performance, whether it is a robot or a live manager.
 
Vladimir Zubov:
I've been getting 100-150% interest on this thing for three years now, but you know where we live, what amounts... and I'm not happy with that.
The amounts can be increased. Can you show the performance in more detail, period by period?
 
Сергей:
Excuse me, but I can't believe you've been trading it for so many years. Otherwise you wouldn't be asking such things. You should not forget about slippages and spread widening without warning, especially on the real market. You may be able to do that on demo. But on the real account, especially if the dealer starts to withdraw you to the interbank ..... If you do not know the difference between the raises and the interbank spread, you may slip and the spread will widen. My personal opinion, I will not waste my time arguing. If you want to discuss it, you should write me in the personal column.
This is exactly what is interesting, this is on the topic of the topic. I said characteristics of really good systems. What they can, how bad differ from the good. For four days we have been talking, but now we are getting down to business.
 
Well, it works for me too)))
 
Oleg Shenker:
It all depends on the number of autocorrelation points and shift range. It may indeed take some time, but you don't have to count it all the time. Autocorrelation is the covariance of a time series with itself with a shifted time lag.

Three months at a time counts. No need to recalculate, the market statistics we need change extremely slowly

Close, but not quite. Corr() and cov() are completely different in form. Just look at plot(corr()) and plot(cov()) and feel the difference.

 
Oleg Shenker:

Nevertheless. I need a general understanding of how the algorithm makes money...

Everything is very simple in fact, remember, as in the old joke: "I buy goods at the wholesale market for a ruble a piece and distribute them among the stalls for three, 3-1=2, that's how I live on these 2%...". But seriously, all trading comes down to trivial buying and selling operations. But it's the principle of determining the right moment for these operations that determines the logic of operation and profitability/loss of the trading algorithm. And the description of this principle in my mind goes beyond the concept of "in general". Therefore, the question arises: "Do you want to drive or drive by car?
Oleg Shenker:

....I don't need your know-how. But, the basic principles of work - by all means, I will not campaign to investors for a black box....

I will not teach you how and what you need to attract and interest investors, but I will risk to express myself on this point with the following analogy - imagine a situation where an employee of a prestigious car dealership wants to sell a client a posh Mercedes or something like that. Would he be interested in consumer attributes such as comfortable leather interior, smooth running and instantaneous acceleration or would he tell them about technology of buffalo hides dressing and technology of engine assembly and production of its components?

Oleg Shenker:

... All the same, I want to understand why the balance staggers so much, but the capital does not. It seems very strange to me. That is, the market value of assets remains constant, but the financial result of closed transactions can fluctuate greatly. To me, it is an obvious sign that the algorithm artificially levels out the equity line, closing either profitable or losing positions. I do not understand why it is done, and if I don't understand it, it means that I cannot use this algorithm...

When you get understanding of it, you can freelance such algorithm for $10. What's my interest then?

Oleg Shenker:

...I think the right thing to do would be to move the conversation to a private room...

Well, I don't mind.

1...1516171819202122232425262728
New comment