Majors CatBoost Clusters Real

Majors CatBoost Clusters Real EA operates on the CatBoost algorithm using clustering methods. CatBoost is fed a set of features based on price and moving averages (MA) .

What exactly:

  1. Distances
    • Price - MA (for different periods)
    • MA − MA
      → show the deviation and strength of the trend
  2. Normalization
    • Division by price or MA
      → makes features scale-independent
  3. Window statistics (WINDOWS)
    • Average, std, max, min
      → reflects volatility and behavior over a period
  4. Lags
    • Past values of attributes
      → give the model dynamics
  5. Relationships (ratio)
    • Price / MA
    • MA / MA
      → relative position
  6. Slope
    • Change in MA over time
      → trend speed
  7. Compression
    • Sum of distances
      → how “compressed” the lines are (flat/trend)
  8. Binary trend indicators
    • MA1 > MA2 > MA3 and vice versa
      → trend structure (up/down)

Recommendations:

  • Connecting an account through the spread rebate service will allow you to earn additional profit in the form of spread rebate.
  • Currency pairs: EURUSD and GBPUSD.
  • Timeframe: H1.
  • The spread is not important, the broker is not important.

Settings:

  • "------------Open settings------------";
  • Intensity - trading intensity;
  • "------------Lots settings------------";
  • MaximumRisk - maximum risk per trade (the number of enabled strategies is taken into account);
  • CustomLot - fixed lot per trade;
  • "------------Close settings------------";
  • TakeProfit - fixed profit;
  • StopLoss - fixed loss;
  • "------------Other settings-----------";
  • MaxSpread - maximum spread for opening a position;
  • Slippage - slippage;
  • Magic - magic number;
  • EAComment - position commentary (ATTENTION - do not change position comments. The EA keeps track of comments).

Altri dall’autore
Majors Perceptron Distance 1
Roman Poshtar
Experts
Majors Perceptron Distance   works on a perceptron recruitment algorithm. The selection and construction of features for training are described in detail in my articles, which you can read below: Experiments with neural networks (Part 1): Revisiting geometry   https://www.mql5.com/en/articles/11077 Experiments with neural networks (Part 2): Smart neural network optimization   https://www.mql5.com/en/articles/11186 Experiments with neural networks (Part 3): Practical application   https://www
Filtro:
Nessuna recensione
Rispondi alla recensione