From theory to practice - page 1435
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
Sash, what tests?
I applied the trades to the sorcerer's indicator, so there is a 50/50 win/loss
Put your parameters for the confidence interval, may be in a personal message.
Exit outside the upper limit - SELL, outside the lower one - BUY? Closed on a pullback and crossing 0?
Tests - give us a start!
OK, mate, you stubbornly defend your point of view. That, of course, deserves respect. But...
Is the market random or not? Let's try to figure it out.
Proof of randomness of the process is that the sum of a large number of independent random numbers belongs to the normal Gaussian distribution.
2. But what happens inside the bars? On ticks?
Well, there the picture is completely different.
The tick distribution does not belong to any known distribution type. I personally did not manage to determine it.
The last thing I came to was that it reminds me a lot of Skellam's distribution used to predict the results of sports games.
How does it work?
Well, it's very simple - if at different moments of time we take the difference of two numbers that belong to two different Poisson distributions, those differences belong to the Skellam distribution.
1. Working with OPEN/CLOSE M1, M5, ... we deal with a random process.
2. When working with ticks inside bars, we deal with a non-random process with its own mathematics, patterns, etc.
That's the tricky mechanism of pricing in the market.
Thank you for your attention.
Hello, something is wrong, but it seems that only in the place where there is a big game, where the small things have no effect, now the high-frequency players rule, and they fight them by making sharp changes in the spread, they even abolish the constant spread everywhere, but because they entered sharply - they got their 10000% in the market (they praised such percentages on the RBC) and left sharply, and when the spread is sharply against them it does not work ...))
Please post the parameters for calculating the confidence interval, you can do so in a private message.
Exit - over the upper limit - SELL, over the lower limit - BUY? Closed on a pullback and crossing 0?
The tests - please, let's have them!
your parameters - not more than the spread, otherwise you will incur losses
Tests - draw them on the chart, I'm not lazy
;)
the parameters - not more than the spread, otherwise you will make a loss
Tests - draw on the chart, I'm not lazy
;)
No, not that... You drew the middle chart correctly, but then you got the wrong one...
Lana, let's wait for Gianni.
Why you?
Obviously, such freaks are necessary in the market - they lower its efficiency.
No, that's not it... You got the middle graph right, and then you got it wrong...
Lana, let's wait for Gianni.
It's not further, it's the end.
Before that, not a few things were done to rule out what you presumably think is right.
Zhenya writes correctly.
This is the problem I have faced.
Yesterday I showed him - all normal, problem solved.
I fixed the time window too.
The lag also resolved.
The redrawing too.
I have no need to worry about ticks, there is no profit there.
On the watch - not more than 2 points.
But beginning from this timeframe and higher we should do the dance, without perversions and pruning.
Otherwise, Zhenya Chumakov, who worked with him, would have become a millionaire long ago. Still, trends (deterministic movements with memory) are tearing him up too.
Or am I wrong? If Zhenya is reading these lines, can he demonstrate the tests?
What is there to demonstrate? We need a filter, and who the hell knows what kind.
Koldun's indicator does not see the trend, so to speak... Zero there is always in a constant place and does not move anywhere... And on the price chart, the average slips... That is why it is sliding ;)
1. Worked with a fixed sliding window as well
2. A window with a fixed point of reference
3. With different dynamic time windows
4. Thought maybe correlation of currencies in a currency pair would help somehow, no.
In short, you need to keep the waiting mat in the same place over time...What's there to demonstrate? That method by itself does not give a profit. You need a filter, but which one?
Koldun's indicator doesn't see the trend, so to speak... Zero there is always in a constant place and does not move anywhere... And on the price chart, the average slides... That is why it is sliding ;)
1. Worked with a fixed sliding window as well
2. A window with a fixed point of reference
3. With different dynamic time windows
4. I thought maybe the correlation of currencies in a currency pair would help somehow, no way.
In short, you need to keep the waiting mat in the same place over time...Are you sure you have the ability and time to run tests on the Warlock indicator?
Need results on any pair with window = 1440. I will send you the correct variance calculation in a private message. No extra parameters, nothing extra. OK?
... The zero there is always in a fixed place and doesn't move anywhere ...
In short, it is necessary that the expectation mat should remain in the same place over time...:)))
Zhenya, please - run some tests. Help an old man out. I'm already wearing rags, eating leftovers... Don't ruin my orthodox soul!