Login / Register
Add a review
Log in
Sign UP
[bws_google_captcha]
Nebula start edition

0.00

Further development of the Expert Advisor on two moving averages https://www.mql5.com/ru/market/product/1693. The Expert Advisor has a self-learning mode that allows you to immediately start trading, but you can also conduct preliminary training of the Expert Advisor for more efficient trading. The Expert Advisor is set to work on the EURUSD timeframe H1 by default.

The Expert Advisor is based on a modified probabilistic neural network. Initially, the neural network does not contain neurons, in this mode, the Expert Advisor does not trade, but only learns. During the work of the Expert Advisor, self-learning takes place and neurons are created in the neural network that are responsible for one of two classes – buy or sell. As soon as at least one neuron appears, the Expert Advisor can trade.

  • Target type – type of target levels:
    • Points – in points;
    • Money-in money.
  • Target profit level, Target loss level, Trailing stop level-setting the target income, loss, and trailing stop levels in points. Applies to the character as a whole;
  • Volume lot – the volume of the trading lot;
  • Slippage – the level of slippage in points;
  • Pause after trades – pause after a trading operation in milliseconds;
  • Number of trade attempt – the number of trade attempts to enter the market. Only applicable for Market mode;
  • Trade mode – mode of trade.
    • Market – open / close positions with market orders;
    • Pending – open / close positions with pending orders.
  • Indicator first period – the period of the first moving average;
  • Indicator second period – the period of the second moving average;
  • Neuron activation level – the level of activation of neurons. The recommended value is at least 0.95;
  • Neural network minimum time signal live, min – the minimum time of the neural network signal in minutes;
  • Neural network renewal period, days – the neural network renewal period in days;
  • Enable load neural network-load neural network data from a file that is required for a safe restart of the Expert Advisor;
  • Prefix file name-setting the file name prefix for storing neural network data. The file is created in the local terminal folder in the FilesNNS directory
  • Enable info panel-display an information panel that displays information about the neural network and trading levels;
  • Expert magic number – unique identifier of the expert;
  • Order comment – a comment on the order.

[spoiler title=”Read More…”]

  • Points – in points;
  • Money-in money.
  • Market – open / close positions with market orders;
  • Pending – open / close positions with pending orders.

Set the Target profit, Target loss, and Trailing Stop parameters to zero. The value of Enable info panel is no. Specify the file name Prefix file name.

Set the optimization values of the two moving average parameters Indicator first period and Indicator second period from 1 to 100 or more. Set the parameter Neural network signal duration, bars from 0 to 10 or more.

Select the optimization period and the testing period, for example, the optimization period is 2016, the testing period is 2017.

Set the optimization mode to 1 minute OHLC. Run the optimization on the period of 2016.

After the optimization is complete, select the appropriate results. When selecting results, follow the following recommendations: the value of Indicator first period should be less than Indicator second period, the number of transactions should be more than 100, and other indicators should be selected at your discretion.

Run a single test of the Expert Advisor with the selected parameters for the period of 2016. If you are satisfied with the obtained indicators, you can check the Expert Advisor for the test period of 2017.

Run a single test on the 2017 test period. If you are satisfied with the test results, the Expert Advisor is ready to trade. The parameters Target profit, Target loss and Trailing Stop can be set at your discretion.

[/spoiler]

Add review
Reviews ( 0 ):