Machine learning in trading: theory, models, practice and algo-trading - page 3068

 
mytarmailS #:

Yeah, I agree, you didn't have a single idea...

all your ideas are to take the last 5 candles and put them in the classifier and all the variations of this...

no thoughts, no creativity, no thinking outside the box... dumb copypaste from overseas blogs of the lowest quality....

That's because you are grey mass and don't understand anything I write. Reflecting your poor perception :)

 
Maxim Dmitrievsky #:

That's because you're a grey mass and don't understand anything I write. Reflecting your poor perception :)

it is better to be a grey mass in the eyes of a sheep than to be a sheep... in my opinion.


Well, launch your mega robot, which of course does not take the last 10 candlesticks as input , and show how it works...

Show the world what a genius you are, why are you making such a big deal about it?

 
mytarmailS #:

it's better to be a grey mass in the eyes of a sheep than to be a sheep... in my opinion.

Well run your mega robot, which of course does not take the last 10 candlesticks as input , and show me how it works....

Show the whole world how brilliant you are, why are you making such a big deal about it...

I don't know what you should be, but if you don't want to talk about MO or fly in with a smart look and wish everyone good luck like your strange R friend, then it's better not to react to new information :)

it was already a donation to those who are hungry for new knowledge, I could have written nothing.

especially if I was offered to use an interesting bot, I would be doubly glad. But you have never posted anything. What is this conversation about :)

 
Maxim Dmitrievsky #:

I don't know what you should be, but if you don't want to talk about MO or fly in with a smart look and wish everyone good luck like your weird R friend, then it's better not to react to new information :)

it was already a donation to those who are hungry for new knowledge, I could have written nothing.

Why don't you want to talk about MO?

Where's the validation? No?

He doesn't have time to do it, come on.

Where's the signal on the demo, no?

No, take my shitty EA and trade, it works, honestly, honestly...


You don't even realise how primitive you are in the eyes of experienced people,

you're just a narcissistic narcissist with a mercurial, boorish, womanising character...

Now I'm done!

 
mytarmailS #:

Why don't you want to talk about the MoD?

Where's the validation? No?

He doesn't have time to do it, come on.

Where's the signal on the demo, no?

No, take my shitty EA and trade, it works, honestly honestly....


You don't even realise how primitive you are in the eyes of experienced people,

a common chatterbox with a mercurial, boorish, womanly character...

now I'm done!

You didn't have to start, it wouldn't have made a difference.

Maybe you're having senile memory lapses. It's been written several times that the OOS is LEFT OF THE PUNCTION LINE.

It was also written that there are VERY many bots, I can collect any, or should I test hundreds of bots specifically for you? Isn't that a bit too much honour?

REMEMBER. LEFT. FROM. PUNCTUAL. LINE.

The rest are your projections of your character onto others.

:))))

you can continue to be an idiot, but you've been intellectually lost to me for a long time :) just for fun. On topic is still not forthcoming.

 
Maxim, can you make a bot on my sample using your system? It would be more interesting to me than just a black box.
 
Maxim Dmitrievsky #:

looks like a Kolmogorov-Smirnov test

in order to test something, you have to first determine what to test. That usually takes forever in our field.

It's desperate.

From the description, it sounds like something else.

"

Description

Finds causal relationships in exact data, detects delays and embeddings in time series, guides the training of neural networks and other smooth models, evaluates their performance, and gives a mathematically sound answer to the overtraining problem. Smooth regression is based on the gamma test, which measures smoothness in multivariate dependence. Causal relationships are smooth, noise is not. "sr" includes the gamma test and search methods that use it.

"

 

I think the issue of causality is important - I wrote about it a couple of months ago with examples of events - but again nobody understood.

The video discusses the importance of causality in analysing the significance of predictors.


 
Aleksey Vyazmikin #:

From the description, it sounds like something else.

"

Description

Finds causal relationships in exact data, detects delays and embeddings in time series, guides the training of neural networks and other smooth models, evaluates their performance, and gives a mathematically sound answer to the overtraining problem. Smooth regression is based on the gamma test, which measures smoothness in multivariate dependence. Causal relationships are smooth, noise is not. "sr" includes the gamma test and search methods that use it.

"

There are many approaches in causal, that was the point, the topic is broad, it makes sense to discuss.
 
Aleksey Vyazmikin #:
Maxim, can you make a bot on my sample using your system? That would be more interesting to me than just a black box.
You must have millions of features in there. My approach is set up to be fully automatic. You can just tell me which features are good in your opinion, several pieces or the same one with different parameters, I can run it on them. And it will pick the targets itself. Because from your dataset only signs will remain in any case, the rest he will change everything else.