JohnLast liked this 4225 days ago
Hi
Here, we are in fact in front of difficult stuff, becoz we used an empiric method to decypher the market
It could be wiser to use fuzzy logic or Swarm Intelligence, to adapt a new method for building a strategy
the question is how ?
I would say we use inductive methods ;), trying everything what we find.
Inductive methods ?
well, the main question is which strategy to use to select the best rules.
I have the book "Trading on the Edge" http://books.google.co.nz/books?id=1XQe-y5sojQC&printsec=frontcover&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false
There is a section that deals with Fuzzy Logic. From memory they had a way of extracting rules from the data. I'll try and find it over the weekend and get more details.
As far as I read the introduction of the book, I thought this was mandatory to have it
the empiric solution to search the best associations of rules, is in fact, very very difficult, even by backtesting
An optimization tool w/ genetic algo, or Neuronal system or, may be Swarm Intelligence should do the work faster
Comments
I would say we use inductive methods ;), trying everything what we find.
Inductive methods ?
well, the main question is which strategy to use to select the best rules.
I have the book "Trading on the Edge" http://books.google.co.nz/books?id=1XQe-y5sojQC&printsec=frontcover&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false
There is a section that deals with Fuzzy Logic. From memory they had a way of extracting rules from the data. I'll try and find it over the weekend and get more details.
As far as I read the introduction of the book, I thought this was mandatory to have it
the empiric solution to search the best associations of rules, is in fact, very very difficult, even by backtesting
An optimization tool w/ genetic algo, or Neuronal system or, may be Swarm Intelligence should do the work faster