The OpenMP/MPICH is to perform parallel optimizations. It really makes no sense to go parallel with the ML training when using this type of algos (the ones I have posted about), it takes milliseconds or less to train them on each bar, it would take ...

Machine Learning with algoTraderJo

Take a look at these results on the USDJPY, using multiple positions as shown here (remember that the ML model is retrained on every bar using the last B examples): SL=80% TP=60% A=22 B=200 C=7 D=60 image This is the first case I present where we ...

Machine Learning with algoTraderJo

You can use many things, those two are excellent python libraries. My main issue with implementing my own framework is that no one helps you debug, so bugs may go unattended for long periods of time. These bugs may cost you a lot of money! I have ...

Machine Learning with algoTraderJo

One of the things I like the most is that you can use the Asirikuy Trader and Strategy Tester programs from Linux. This is really a must for me, to be able to live trade and backtest on Linux. The strategy tester also supports OpenMP/MPICH which is ...

Machine Learning with algoTraderJo

There are several front-ends you can use. You can trade/backtest the systems in MT4/MT5 or in the programs developed by the community (Asirikuy Trader and Strategy Tester). Using the Asirikuy Trader you can live trade using the Oanda or Dukascopy ...

Machine Learning with algoTraderJo

Some USDJPY results using the same modifications as in this post : SL=90% TP=350% A=12 B=50 C=6 D=180 image This is just a simple example after a small search... much more linear results are indeed possible... This machine learning algo can create ...

Machine Learning with algoTraderJo

Systems in the F4 framework are coded in C/C++ if you want to code using the framework you naturally need to know these languages.

Machine Learning with algoTraderJo

A couple of Asirikuy members are helping me double check the functions before they are added to the F4 unstable branch this week (I think). I will post them when they are ready. I don't want to post messy code here.

Machine Learning with algoTraderJo

You can also modify the trading system in the following ways: Take a take long trade whenever the long prediction is positive (value of the short prediction does not matter) Take a short trade whenever the short prediction is positive (value of the ...

Machine Learning with algoTraderJo

If you want to compare you can do so by looking at my previous posts. I don't have enough time available to help you with this task. Don't ask me about my preferences. I am just showing you different machine learning approaches, it is up to you to ...

Machine Learning with algoTraderJo

This means that the example is not valid so it is skipped.

Machine Learning with algoTraderJo

The EU and GU results can be combined in a portfolio to give better performance than the individual systems: image

Machine Learning with algoTraderJo

I always use GMT +1/+2 this is extremely important, use GMT with the wrong shift or without DST and you won't be able to reproduce. the input is the return of the past 9 bars defined as (Open[N]-Open[N-1])/Open[N-1] where N=0 is the currently used ...

Machine Learning with algoTraderJo

Remember that these systems work on the 1H chart and also remember that I am not using MSE/MLE here but a new algo (read this post carefully url ). This is also a linear regression algorithm, if you want to reproduce avoid using classifiers.

Machine Learning with algoTraderJo

We can also get good linear results (correlation coefficient = 0.96) on the GBPUSD: SL=70% TP=220% A=14 B=225 C=9 D=70 image

Machine Learning with algoTraderJo

Precisely the point. Another question, is including data up-to-present always the best thing to do?

Machine Learning with algoTraderJo

No I changed the technique since I restarted posting, read this post for more details about the last ML algos http://www.forexfactory.com/showthread.php?p=8417689#post8417689.

Machine Learning with algoTraderJo

I don't think I agree with "the longer you wait the less confident you are about which of the SL or the TP comes first". When I create the examples used for the building of the training dataset on each bar I only care that the SL or TP is hit within ...

Machine Learning with algoTraderJo

We can also get highly linear results (correlation coefficient > 0.94) using this type of strategy. SL=40% TP=280% A=10 B=75 C=11 D=150 image

Machine Learning with algoTraderJo

1. Depends, I only ensure that results do not change very significantly on the broker where I want to live trade. I will generally perform a back-test in Oanda Vs 1987-2015 data using Oanda data from 2004 and in this way I'll check that results do ...

Machine Learning with algoTraderJo