Machine learning in trading: theory, models, practice and algo-trading - page 1320
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
Doc is most likely alive, just offended by numero uno and the fact that he drained bitcoins on Bitrex, but Alyosha and Jura are most likely really in the netherworld, and were probably there under violent circumstances. Here's an advertisement for Forex... It would be in theme, if somebody would sing something about how the market rises and breaks the guys, both stupid and smart, both cowards and desperate daredevils, something like Makarevich's "One day the world will bend for us".
Anything can happen. No one lives forever under the moon.
It's a shame Alyosha deleted his posts. No Alesha, no his masterpiece posts, imbued with love for neuroGraal, nurtured in home conditions... Ehhh... That brings a tear to my eye.
Anything can happen. No one lives forever under the moon.
It's a pity that Alyosha deleted his posts. No Alyosha, no his masterpiece posts, imbued with love for homegrown neuroGraal... Ehhh... I'm on the verge of tears...
Words are not physics, but lyrics, probably that is why your calls to the community are full of expression and poetry, but not motivation and technical sense.
For example, with BP memory analysis, section allocation, etc., for such work you need a normal problem statement, not a link to someone's "undigested" files in an internal format with no author to ask something and call to guess something there.
At best, this may bait some adventurer, who will just say that he has already guessed everything and promote his screenshots with indicators or balance charts on the occasion.
But if this is not the goal, then offer normal input data, for example CSV files for uploading to custom MT5 symbols with additional hidden sets of forward tests on which you can check the results, formulate a clear, unambiguous task and organize the work, such as a test or as a competition of models of MO.
I think people's penchant for squeamishness and your talent for sparking may well melt the local community:)
Alexey chose not the worst catbust, one of the best for research, today. It is used at CERN, for example, to analyze collider results... who deals with quantum randomness.)
More details here, please. Catbust is a product of Yandex. Or I am mistaken?
More details here, please. Catbust, I think, is a product of Yandex. Or am I wrong?
Does it contradict something?
Here's the problem from Doc again
His model was able to solve the problem and made a profit on one of the artificial rows attached to that post. If your model can't do that - maybe it's too early to take on real BPs?
He was working with first differences - increments on thinned real BPs, and something else (he didn't tell me everything, of course). His model predicted the sign of the next increment and his signal was positive.
So why does everyone here spit on the experience of successful traders and reinvent everything here, scaring the youth with some herbarium! I do not understand. I refuse to understand.
The memory is good.
But it melts over time.
It's the same on the market.
So the task is even more difficult.
It is solvable, of course.
Does it contradict anything?
Should it? Maybe I just did not understand what we are talking about, I want to clarify.
Should it? Maybe I just do not understand what we are talking about, I want to clarify.
Clarify what? Yandex product yes.
:))) I'll think about it.
In the meantime, once again, what I've actually seen working on neuronics:
1. increments on a thinned teak row are used.
2. Thinning is carried out on a cunning algorithm: a) the difference between two consecutive values should exceed the spread, b) the density of the distribution of increments should be maximally symmetric, i.e. the asymmetry coefficient should be =0 for any sample.
3. The autocorrelation coefficient of the increments is also important.
4. Predicts the sign of the next increment
As you can see, the most difficult task is number 2. It's the one that needs to be given unrestrained attention.
Everything else is crap and nonsense.
:))) I'll think about it.
In the meantime, once again, what I've actually seen working on neuronics:
1. increments on a thinned tick row are used.
2. Thinning is carried out on a cunning algorithm: a) the difference between two consecutive values should exceed the spread, b) the density of the distribution of increments should be maximally symmetric, i.e. the asymmetry coefficient should be =0 for any sample.
3. The autocorrelation coefficient of the increments is also important.
4. Predicts the sign of the next increment
As you can see, the most difficult task is number 2. It's the one that needs to be given unrestrained attention.
Everything else is crap and nonsense.
Is it about points other than the second or about my today's real account?
Get out of here.
Uncle, I didn't write you
Don't interfere where you're not asked, or be tactful in your communication
or I'll press the "Violation" button again.