Discussion of article "Experiments with neural networks (Part 1): Revisiting geometry"

 

New article Experiments with neural networks (Part 1): Revisiting geometry has been published:

In this article, I will use experimentation and non-standard approaches to develop a profitable trading system and check whether neural networks can be of any help for traders.

Optimization and forward test results.

4 perceptron 4 tangent

Forward test date from 2021.05.31 to 2022.05.31. Out of all the results, we should choose the one featuring the largest profit factor with the maximum of the complex criterion exceeding 20-40.

Test 1

Test 2

Author: Roman Poshtar

 
Nice insight. I also try experimenting with neural network and have come across similar idea, not passing the prices directly to the network as prices have no valid boundaries thus I choose to pass the angle/slope of the moving averages.
Let me know if you found some progress.
 

Hi Roman, thank you for your idea and sharing, may I ask for a query about the Perception4:

double perceptront4() 
  {
   double w1 = x1 - 100.0;
   double w2 = x2 - 100.0;
   double w3 = x3 - 100.0;
   double w4 = x4 - 100.0;
   
   double a1 = (ind_In1[1]-ind_In1[10])/10;
   double a2 = (ind_In2[1]-ind_In1[10])/10;
   double a3 = (ind_In1[1]-ind_In1[10])/10;
   double a4 = (ind_In2[1]-ind_In2[10])/10;
   
   return (w1 * a1 + w2 * a2 + w3 * a3 + w4 * a4);
  }

according to your code, it seems a3 mis-typed equal to a1 = (ind_In1[1]-ind_In1[10])/10;

(I found the source code is the same as be)

Refer to the figure (as attached), (I think) it should be

   double a3 = (ind_In1[1]-ind_In2[10])/10;


I'm a newer to MT4/5 and neural network, it is an very interesting to study and discover your AI method article, is there any chance to change your results and conclusion? 

hope to see your next article.


Best regards

WK

Files:
 
День семьи ФФДень10 # :

Привет, Роман, спасибо за идею и то, что поделились, могу я задать вопрос о Perception4:

согласно вашему коду, кажется, что a3 неправильно выбрано, равно a1 = (ind_In1[1]-ind_In1[10])/10;

(Я наблюдал, что исходный код такой же, как и был)

Обратитесь к рисунку (как прилагается), (я думаю) он должен быть


Я новичок в MT4/5 и  нейронной сети , очень интересно узнать и открыть для себя вашу статью по методу ИИ, есть ли шанс изменить ваши результаты и заключение? 

надеюсь увидеть вашу статью.


с настройками

ВК

Hello. Perhaps there is a typo. Try to change. Thanks for your feedback.