You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Hi all!
I am a mql4 programmer, I have already worked into NN EA's. I don't need to speak about me, rather about a project's of mine for active minds.
I have a matrix of hundreds of samples of three inputs for an output from eur usd M5 history.
Inputs are particular definitions of the properties of a bunch of ma's respect to the price everywhen a certain event (pattern) occurred.
Output is correspondently the price change net of a maximal of adversity tollerance I set.
Assuming those data being really well done and constituiting an excellent way to see forecasting problems, it would be very interesting to elaborate algorithms to model the most reliable, stable and profitable relation(s) between those in and out.
Ermeneutically, there could be at least 2 ideally separate universes of techniques:
- 1) aiming at solve the problem by finding those weights that can approximate the least erroneus relations (nonlinear NN and Regression Worlds);
- 2) aiming at rethink the problem by finding out that submatrix characterized by the strongest inner relations so that to apply modeling much more effectively (linear/nonlinear Statistical Analysis, Cluster Analysis, Pattern Analysis/Classification Worlds);
- 3) whatever you can figure out differentely from the 2 ways above.
Tools for experiments are very famous amongst mathematicians and quite spreaded on the internet.
I can provide the csv files with already scaled (and clamped) inputs/output, ready to be analyzed.
Final task could be that of Creating Together an EA.
Any expert helps is the welcome. Could come from any of the subject matters and contact me at minkobliss@hotmail.it. Please introduce yourself, why interested in that and what the frameworks and background theories of your approach.
Thanks for your attention.
OF COURSE:
- I am going NOT TO PAY anyone in case. I spent my life to understand what does matter amongst all the price action patterns/indicators/oscillators in the world. I choose inputs and output very carefully, the script to extract them is, alone, 1k lines.
- Everything could emerge from those analysis MUST BE SHARED. It is not a get-and-run away scheme.
- BPNN is not enough for that issue.
Thanks again.
SimonTrader