• 概要
  • レビュー
  • コメント
  • 最新情報

Dopamine Pro

Trading algorithm based on technologies similar to Google Deep Mind (AI based). Easy to set up and learn.

It works on any financial instruments and any time frames, as you wish. It is possible to create a basket of strategies by training it several times with different Magic position numbers.

Advisor settings:

Optimization settings

  • Order magic             - a unique magic of positions, is also a counter for optimization
  • Learning mode  flag setting the training mode. For trading or testing should be disabled
  • Period_TF                - the opening interval of transactions, a signal will appear every bar of the selected time frame.
  • Symbols List             list of analyzed instruments, separated by commas
  • features_number      depth of analyzed history
  • best_features           - number of best selected features
  • Trades frequency      - the frequency of transactions for training, from 0 to 1, the higher the more transactions, it is recommended 0.1 - 0.9
Position control
  • Maximum risk     - progressive lot management system, the higher the value, the greater the position volume
  • Custom lot         - fixed volume trading, if 0 then progressive is used
  • Stop loss            - protective stop loss for each position
  • Take_profit
Fine-tuning
  • Decision boundary  - the decision boundary, in the range from 0 to 0.5, the higher the value, the greater the accuracy, but fewer transactions (and vice versa)

Before use, the expert must be trained on a specific currency pair and time frame!

After you have downloaded the expert, you need to run it in the strategy optimizer , according to the " Custom max " criterion , with an exhaustive search algorithm, for this, additionally:
  • Set the range " Order Magic" in the optimizer, for example, start: 1, step: 1, stop: 10
  • Set the " Trades frequency " parameter to the desired number of trades. This parameter is used only during optimization to select a strategy, and does not participate in real trading.
  • After waiting for the optimization to finish, run the robot in the tester, increasing the test period to see how the adviser is trading on new data. Adjust the " Decision boundary " parameter, which will get rid of too frequent transactions and improve the performance of the system. Set " Custom lot" or " Maximum risk" to an acceptable value for you, check again in the tester.
  • After that, you can save the set and load the system onto the chart, using the saved set, or set the settings manually.
You can do this several times by changing Order magic or time frames or financial instruments, thereby creating an unlimited number of strategies and diversifying risks.

Do not train an expert for too long sections of history: this can take a long time. Focus on testing and evaluating trading on new data.

Your dopamine in the world of strategies.


おすすめの製品
フィルタ:
レビューなし
バージョン 2.3 2020.06.20
Added the ability to optimize stop loss and take profit
バージョン 2.0 2020.06.18
There was an opportunity of optimization among trained models.