Gathering a team to develop an IO (decision tree/forest) in relation to trend strategies
I propose to spar for solving an MO problem in relation to trends, i.e. when the probability of an outcome event should
3. lisa with knowledge of iO and/or statistics
Getting burned at work is not our method. )))
Women for statistics ))))
Getting burnt on the job is not our method. )))
Women for the statistics ))))
Thank you for the comments with humour! Yes, I have that flaw - I make mistakes.
A little bit about the directions I see as promising:
1. Work on genetic methods - here I want to try to create logical groups of features, including sowing and nurturing offspring.
2. Work on fitness functions - here I want to provide feedback to actual trade results, not just classification results.
3. Work on stumpage nurturing - a semi-automatic method of generating trees when the base of the tree is known or when individual branches are known.
4. Work on a method for evaluating the resulting tree.
5. Work on a convenient and fast way to integrate the tree into the EA algorithm.
6. Working on creating a herbarium - a method of pulling out individual rules to create forests.
7. Work on method of tree building not excluding limits/ranges, but trying to give solutions to all values in the predictor (relevant when quantization (subdivision into subgroups) is already used in preparation of the predictor).
Without a description of tools and version control, apart from the desire, there is nothing here yet...
What do you mean by tool descriptions and version control?
I do not know the name of libraries and specific algorithms, I have ideas on paper and partly in code, but I am ready to discuss it with interested persons - this is a creative project!
What do you mean by tool descriptions and version control?
I don't know the name of libraries and specific algorithms, I have ideas on paper and partly in code, but I'm willing to discuss it with interested people - it's a creative project!
You put R (15 min), in R you put the rattle package, another 5 min. This package has raw data preparation, 6 models including tree and random forest, and a fairly extensive model evaluation unit.
Loading input data from excel, you can unload a bunch of different predictors from the Expert Advisor to excel: different currency pairs, indicators...
All this you may run it for at least a week. You will get ready models with performance estimation, with charts and stuff.
After that you will be able to formulate your proposals on some minimal professional level.
Put in R (15 min), in R put in the rattle package, another 5 min. This package has preparation of initial data, 6 models, including tree and random forest, and a rather detailed block of model evaluation.
Loading input data from excel, you can unload a bunch of different predictors from the Expert Advisor to excel: different currency pairs, indicators...
All this you are running for at least a week. You will get ready models with performance estimation, with charts and stuff.
After that you will be able to formulate your proposals at a minimum professional level.
What task is not clear to you? Why are you criticizing? Don't you remember that I put Rattle, looked and felt it. I have concluded that the standard approach is no good and a unique method for trading is required. I'm surprised you haven't realised this over the years.
Hi!
Putting together a team for Forest and Trees?
What's that?
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
I propose to rally to the challenge of MO in relation to trends, i.e. where the probability of an outcome event must be predicted many bars ahead.
What I propose from my side:
1. a basic strategy for generating targets.
2. Maintaining a joint database of features.
3. Organizing the possibility of optimization/other calculations which require computer power. If necessary, capacity expansion. Possible creation of a common network or optimisation on behalf of.
4. Conducting tests in the strategy tester in order to detect errors in the algorithm's logic.
5. Testing of strategies on demo accounts/cent accounts/other accounts with small lots. If necessary, I will provide a machine for VPS (though I will need help in setting).
6. Holding collective meetings to develop common goals and activities.
7. Providing ideas for building a decision tree.
8. Coordinating the work of the team.
Who is needed:
1. Programmers able to work with MQL5
2. Programmers with skills in programming languages like R and/or Phyton.
3. people with knowledge of MO and/or statistics 4.
Other persons willing to contribute to the project and ready to offer something - the question is negotiable.
About me - I trade on Forex since 2006 with small breaks, on MOEX since 2016. I have experience in building exchanges. I have a higher education - I graduated from economics department with specialization in accounting, analysis and audit. Currently I do not have a steady job - I am focusing only on the financial markets, or rather on the creation of ATSs. I plan to receive the certificate of the professional participant of a securities market (if who wants with me in a marriage to pay grants for formation - write in a personal - we shall discuss).
If you have any questions - ask, for now I can show you the results in the test of my approach to create an ATS on the decision tree outside the period of training:
Of course the report is not perfect and there are many ways to straighten it out, I just want to show the potential of the approach.