AI Trading Strategy Using Python and MT5 Without Traditional Indicators

I am looking for a skilled programmer to develop the following strategy:

1. Connecting Python to MT5:

   • Integrate Python with the MT5 platform and interact with the trading account in real-time using the following login credentials:
     • account_id
     • password
     • server

2. Using Advanced Python Libraries:

   • Utilize advanced Python libraries to design and train neural networks that predict market trends.
   • Rely on real-time market data (such as previous prices and historical data for both currencies and gold).

3. Historical Data:

   • Fetch historical data from the MT5 platform for currencies and gold (such as close, open, high, low prices, and volume) for various timeframes (1 minute, 5 minutes, 15 minutes, hourly, daily, weekly, monthly).
   • Split the data into training and testing sets, continuously updating the data while the system is running to ensure the model works in real-time.

4. Neural Network Design:

   • Design a Multi-layer Perceptron (MLP) or use advanced prediction models like LSTM or GRU to handle time series data and forecast future market trends.
   • The neural network should either be an MLP, LSTM, or GRU network to process time-series data.
   • Use advanced models like LSTM or GRU, which are capable of handling time-series data and predicting future market direction based on previous price patterns.
   • The neural network should learn to recognize repeated patterns in price movements, such as large spikes or reversal patterns, without relying on traditional indicators.

5. Training the Model:

   • Train the model on data such as:
     • Series of open and close prices.
     • Highs and lows over different timeframes.

6. Focusing on Raw Price Patterns:

   • Design a neural network that focuses on learning hidden price patterns in raw data without relying on technical indicators.
   • Use raw price data (such as open, close, high, low prices, and volume) instead of technical indicators.
   • Prepare the data to be clean and usable for deep prediction models, with continuous updates in real-time.

7. Making Trading Decisions:

   • Make buy or sell decisions based on the model’s output, using Stop Loss (SL) and Take Profit (TP) levels determined by the model's analysis and potential risks.
   • For example:
     • If the neural network predicts a price increase with a confidence level above 75%, a buy trade is opened.
     • If the neural network predicts a price decrease with a confidence level above 75%, a sell trade is opened.
   • Stop Loss (SL) and Take Profit (TP) levels are set based on the model’s output and the level of potential risk.

8. Risk Management:

   • Determine the lot size (trade size) based on the account balance and the defined risk level (for example, risking 1-2% of the capital per trade).
   • Use a Trailing Stop to protect profits in case the market moves in favor of the trade.
   • Monitor all open trades and automatically close them if the prediction changes or targets are reached.

9. Direct Execution:

   • Use the MetaTrader5 library to send orders directly to the MT5 platform (open trades, set Stop Loss, and Take Profit).
   • Continuously update predictions and signals based on new market movements and updates from the neural network.

This strategy aims to execute trades directly in the MT5 platform based on real-time market predictions from a neural network, without the use of traditional indicators.