Machine learning in trading: theory, models, practice and algo-trading - page 2173

New comment
 
Maxim Dmitrievsky:

the topic got serious )

here's a more complicated (but promising) way to read

https://ml4trading.io/chapter/19

This is about densities and how you can sample features and labels from an posterior using an autoencoder. You can also transpose feature spaces, to your liking.

CVAE (conditinal variational autoencoders) in particular are interesting

Thank you very much! I've tried to use the principles of autoencoding with simple full-knowledge neural networks, but it was not entirely successful))) Maybe I should reconsider the experience))

 

Thank you very much! I tried to use the principles of autoencoding with simple full-knowledge neural networks, but not quite successfully.) Maybe I should reconsider the experience))

I wrote an article with an example on Gaussian mixes. I haven't got the encoder figured out yet, then I want to replace the mixes with the encoder

If they publish it, you can read it

 
Evgeniy Chumakov:


You were right to characterize yourself, it's quite a ...


thank you, now everyone knows that

that a person (i.e. you) is very tactful in sucking the meaning of my system and then bang and...

instead of thank you he puts up a post that makes you want to stand on your feet

By the way, the photo shows me building my first oil bush
 
Maxim Dmitrievsky:

I wrote an article with an example on Gaussian mixes. I haven't figured out the encoder yet, then I want to replace the mixes with an encoder

If they publish it you can read it

By all means, I read your articles since my experiments with fuzzy logic. Based on your ideas in this thread I wrote a random forest parser from python to mql5. You had sparred catbust by then)))
 
Renat Akhtyamov:

Thank you, now everyone knows that.

that a man sucks the sense of the system very tactfully and then bam and...


The key word is man, and you're ga...

 
Evgeniy Chumakov:

The key word is human, and you're a sh...

man, it's easy just to praise

any luck?

 
Renat Akhtyamov:

You're 67% of the balance

ahahaha

what's the slag in your monitoring?

I don't learn on bitcoin, I'm too lazy to look for the error. I don't know if the spreads are tight in roboforex or what

I also have to use eurodoll.

 
Alexander_K:

Yeah...

The subject gets better every day.

And, after all, it seems to be simple. There are coryphaei of MO here, but practically, there are no wizards who can offer some magical features (right?), which would have predictive ability. Proving theoretically that these are really great chips, of course...

Now, it has been said, not by me, but by everyone, that the simplest model for prediction is the sign of the next return of a stationary process with a non-zero ACF.

The ACF is lucky - in the market, no matter what you do with the quotes, it is always nonzero and this gives a wild hope in the hearts of those who suffer.

But where is stationarity and how can we predict the sign of the next increment if it may be "stupid" = 0?

Ahem... I've already shown a passage from the Book of Genesis many times. I'll show it again:

"... everything is elementary, the stationary distribution is obtained by writing bars with equal number of ticks.
at that the distribution of time intervals between OPEN of such bars is exponential:

, and the distribution of increments is double triangular:

Now, having such a distribution of returns, predicting the sign of the next one does not seem to be such an impossible task.

Ahem...

Beautiful damn words. I like them. Hee hee.

 
Evgeniy Chumakov:

- Blue Eyed Cote - 'I found another error in the formula all now grail > run monitoring > equity down > delete monitoring > another error > new rena . > formula

- Sweaty Musician - 'OK from Monday I'll run the signal to show how cool I am > equity down > removing monitoring > this is different ... > that's the other thing I was testing.

Sounds like you had a beer, brother. And beer is bad for your thinking centers.

 
Maxim Dmitrievsky:

Throwing a bot for whiners or just interested:

Train + validation from 2020.06

Test (model selection) from 2020.01

additional test from 2018. (earlier periods do not pass)

H1 TIMEFRAME

You may use demo order.

I can train better, but I am lazy today. This should definitely be enough for a week of trading.

I have learned it on several MAs. It is not a grail, but it demonstrates the capabilities of the MO.

It's just that the conditions are steep, I've already said that.

i'm not going to throw the post where you blabbed about spread on bitcoin at you, i forgive you ;)

the commission is 6 times the spread

not any system will fit there

;)))

well, so be it ava

for the sake of decorum, I'll at least start it on cents, so that the computer does not crap on the market, let's say ;)))

But you can run it on a demo, I don't care.

1...216621672168216921702171217221732174217521762177217821792180...3680
New comment