Discussion of article "Machine Learning: How Support Vector Machines can be used in Trading" - page 2

New comment
 
Very good. Good thinking.
 
great
 

Hi

This article is very high quality. Thank you for the good work.

However, I do have a question or challenge for the author:

According to Dr Ernest Chan, using machine learning to predict price movement and drive a trading algorithm is extremely difficult and generally yields poor results. It seems trying to generate alpha in this way is only profitable for an elite of hedge funds. According to Dr Chan, machine learning can be used much more effectively by anyone to predict the probability of success of a strategy given the current market conditions. 

See this playlist: Financial Machine Learning Course | PredictNow.ai - YouTube

In particular, see this video contrasting these two ways to apply ML to trading:  The Two Methods for Using Machine Learning in Trading | Financial Machine Learning Course - YouTube

--------------

The question for you: 

Have you tried both ways of using ML to trading? If so, what is your experience with both?

In your experience, can one generate alpha using ML to make price change predictions and generate trading signals, directly implementing the trading strategy?

 

hello, thanks for sharing

 

Hi, Thaanks for the article. I tried to run your code to see a live example using the code you added and it gave me 4 error's


'OP_TOLERANCE' - undeclared identifier  Line 52

'OP_TOLERANCE' - cannot convert enum Line 52

void setParameter(int,ENUM_OPTION,double)               Line 23

'OP_TOLERANCE' - undeclared identifier  Line 60

'OP_TOLERANCE' - cannot convert enum Line 60

void setParameter(int,ENUM_OPTION,double)    Line 23


I get this on both schnick.mq5 & Schnick_demo.mq5 


Any suggestionson resolving these errors?

12
New comment