08. Case Study: Introduction

AI For Trading C6 L3 A05 Case Study- Introduction V2

Overview of AI Trading Model Setup

This section provides a practical framework for setting up an AI trading model using a momentum-based strategy enhanced with machine learning. Here's a step-by-step guide:

  1. Define the Strategy:

    • Use Moving Average (MA) Crossover for bullish and bearish signals.
    • Enhance with a binary classification model to predict buy/sell opportunities.

  2. Collect Raw Data:

    • Historical data: open, low, high, close, volume.
    • Include a volatility index like VIX.

  3. Feature Engineering:

    • Indicators: Lagged returns, short and long-term moving averages.
    • Signals: Use crossover strategy as a ternary variable (1, -1, 0).
    • Additional options: MACD, Bollinger Bands, RSI.

  4. Feature Scaling:

    • Methods: Standardization, normalization.

  5. Define Prediction Target:

    • Binary classification for profitable buys.
    • Profitability threshold and time horizon as hyperparameters.

  6. Data Splitting & Validation:

    • Test/train ratio, cross-validation with time-based methods.

  7. Model Selection:

    • Choose model and identify related hyperparameters.

Next, focus on hyperparameter tuning.

Next Concept