The philosophy of algotrading - page 16

 
Реter Konow:

1. however - this is the only tool and the only approach used by most scalpers. When you say there are no stable profitable EAs these days - don't you mean scalping EAs? Doesn't it demonstrate the effectiveness of mathematical prediction of price used by scalpers?

2. Let me try to use a simple analogy to show an example that will explain my disbelief in mathematical price prediction:

Imagine a theatrical stage. The stage. The curtain is closed, but the action is going on...


1. Why are you twisting things around like that? Did I say anything about scalping? Quite the opposite - I wanted to bring to your attention that patterns in scalping do not work because of the great information noise...

By the way, the example ( see above) refers to medium-term strategies ( if you don't get it )...

The efficiency of mathematical forecasting in scalping is very low...


2. Example for those who do not understand thechanalysis, and here I agree with you, they are deaf!!!


The conclusion is simple - you have too much knowledge of theory, but no practical knowledge of trading.... This needs to be corrected, and as soon as possible...

 
Serqey Nikitin:
...


The conclusion is simple - you have too much knowledge of theory, but no practical knowledge of trading.... This needs to be corrected, and as soon as possible...

You are right here. I have very little personal experience in trading, but I judge by other people's experience...

My current challenge, is to improve the level of trading robots in everything, namely:

  • Robots must have an interface.
  • Robots should have a minimum set of automated technical analysis features.
  • Robots should analyze a greater variety of data.

The last point is one of the most important in my concept.

Well, I see the future of trading robots in their striving for artificial intelligence. The current trend of development - interface-free robots analyzing only price and its derivatives in different ways is a dead end.

Robots need to be made smarter, more sophisticated and expand the scope of their interaction with the user.

Most importantly, such changes in robots will change the nature of the market itself and its infrastructure in the future.

That's the way to go...

 
Serqey Nikitin:

1. Why are you twisting things around like that? Did I say anywhere that I was talking about scalping? On the contrary - I wanted to let you know that patterns in scalping don't work because of a lot of information noise...

By the way, the example ( see above) refers to medium-term strategies ( if you don't get it )...

The efficiency of mathematical forecasting in scalping is very low...


2. Example for those who do not understand thechanalysis, and here I agree with you, they are deaf!!!


The conclusion is simple - you have too much knowledge of theory, but no practical knowledge of trading.... This needs to be corrected, and as soon as possible...


You're right here. I have very little personal experience in trading, but I judge by other people's experience...

My current challenge, is to improve the level of trading robots in everything, namely:

  • Robots must have an interface.
  • Robots should have a minimum set of automatic technical analysis features.
  • Robots should analyze a greater variety of data.

The last point is one of the most important in my concept.

I see the future of trading robots in their striving for artificial intelligence. The current trend of development - interfaceless robots that analyze only price and its derivatives in different ways, I think that it is a dead end.

Robots need to be made smarter, more sophisticated and expand the scope of their interaction with the user.

Most importantly, such changes in robots will change the nature of the market itself and its infrastructure in the future.

That's the way to go...
 
Реter Konow:

Here you are right. I have very little personal experience in trading, but I judge by other people's experience...

My current objective, is to improve the level of trading robots in everything, namely:

  • Robots must have an interface.
  • Robots should have a minimum set of automatic technical analysis features.
  • Robots should analyze a greater variety of data.

The last point is one of the most important in my concept.

Well, I see the future of trading robots in their pursuit of artificial intelligence. In my opinion, the current direction of development - interfaceless robots that analyze only price and its derivatives in a different way is a dead end.

Robots need to be made smarter, more sophisticated and expand the scope of their interaction with the user.

Most importantly, such changes in robots will change the nature of the market itself and its infrastructure in the future.

That's the way to go...

All this is correct!!! The directions are obvious, but not at the present time...

There has to be a BASE of stable profitable robots, and only then can the issues on your list be addressed.

For now - this is just a wish for the future development of trading ...


Tip: Start by drawing up such a BASE...


 
Serqey Nikitin:

That's all right!!! The directions are obvious, but not at the moment...

There has to be a BASE of stable profitable robots, and only then can the issues on your list be addressed.

In the meantime - these are only wishes for the future development of trading...


Tip: Start by compiling such a BASE...


I will try to prepare a basis for a fundamental upgrade of modern robots. Give them an interface and a couple of other effective tools (which will take another couple of years to develop).

And then, - this base will be developed by community developers.

It will be up to them.

Once again I repeat my opinion - profitability and scalping are incompatible.

Profitability is possible in wise trading - in a strategy that takes into account a lot of data and analyzes the market in different time scales.

Because of the current imperfection of Expert Advisors, the full automation can only be unprofitable, so to improve the profitability, you need to change the current trend in preferred strategies and analysis tools, and in no case completely exclude human participation in the trading process.

The very concept of fully automated trading should not be popularized yet, because robots are not yet very effective, and their profitability without human involvement may be random and short-lived.

P/S For the advice, thanks. )

Reason: