Machine learning in trading: theory, models, practice and algo-trading - page 1022

 
Igor Makanu:

My opinion is that there has never been any average strategy in the markets and there never will be, well, it's like a game without rules:

We were given a field and two colored chips, but the rules were not explained, you thought we were playing checkers, I thought we were playing backgammon, and then a market maker, who plays Chapaev, showed up and scattered all our chips

))))

Already posted somewhere that the Central Bank of Russia, and I think in other countries, the practice of classical technical analysis, and therefore one should expect that most of the participants, not even by choice but because of their training (prof. attestation), will use this notorious technical analysis. I do not understand how it can be denied.

 
Aleksey Vyazmikin:

Already posted somewhere that the Central Bank of Russia, and I think other countries, the practice of classical technical analysis, and therefore we should expect that most of the participants, not even by choice but by the fact of their training (professional certification), will use this notorious technical analysis. I do not understand how this can be denied.

I think it's true - watch on YouTube how the chairman of the Central Bank deputies Chehvosti, the results of foreign exchange transactions of the Central Bank)))

 

Yoo-hoo bros!!! I left you behind so the thread immediately fell below the first page. Yeah.... i didn't expect this from you, i didn't expect this from you.....

Me??? I'm doing fine.... I'm working in production..... and continuing to teach the robot all sorts of wisdom..... What else is new... Optimized JPrediction for my needs, models are getting even better and more adequate to the market. Trading is going slowly..... All new trades are tested by the battle, so capital gains are lame.....

So it is unlikely to work with the video, which I promised.... Although the summer is not over and September is still ahead, maybe something will happen.

So, like this.... So, how are you guys here? Who's got what, if there's an opening .... or had a breakthrough. Share your joys....

 
Mihail Marchukajtes:

Anyway, it's like this.... Well, how about you guys here? Who's got what, if there's an opening .... or had a breakthrough. Share your joys....

No discoveries. Everything is open before the New Year.)
I'm in a state of confusion about what to do. No new thoughts. So far I have not found anything worthy in my posts.
 
Yuriy Asaulenko:
No discoveries. Everything is open before New Year.)
Now I'm in a state of confusion, wondering what to do. No new thoughts. So far I have not found anything worthwhile in my posts.

This is a temporary illusion.

You do not understand why one and the same strategy behaves differently in different markets
 
Renat Akhtyamov:

This is a temporary illusion.

It's not clear to you why the same strategy behaves differently in different markets
No, it is not clear. Any adequate strategy earns in any market. Tuning and profitability is another question.
 

Two weeks ago I created an indicator that builds levels, I am still just learning to understand it and create a trading system based on it, that's why trades are sloppy and I do not always manage to take a trend.

They work out with unknown for me frequency (+-70%).

The indicator is based on neural networks

Several pictures

I've realized for myself and I've already told about it here many times...

The market is not a time series, the market is levels and events, but not the time sequence of points

Good to all

 
mytarmailS:

Two weeks ago I created an indicator that builds levels, I am still just learning to understand it and create a trading system based on it, that's why trades are sloppy and I don't always manage to take a trend.

They work out with unknown for me frequency (+-70%).

The indicator is based on neural networks

Several pictures

I've realized for myself and I've already told about it here many times...

The market is not a time series, the market is levels and events, but not the time sequence of points

Good luck to all

You don't need neural networks to build levels.

 
mytarmailS

The market is not a time series, the market is levels and events but not a time sequence of points

Good luck to all

There's not much left to do but give up levels.)) The market is only events). The rest is secondary.
 

https://shapeofdata.wordpress.com/2013/07/23/gaussian-kernels/

Especially for Yuri, who has nothing better to do

Gaussian kernels
Gaussian kernels
  • Jesse Johnson
  • shapeofdata.wordpress.com
In order to give a proper introduction to Gaussian kernels, this week’s post is going to start out a little bit more abstract than usual. This level of abstraction isn’t strictly necessary to understand how Gaussian kernels work, but the abstract perspective can be extremely useful as a source of intuition when trying to understand probability...