Yo-Yo.ai Docs
  • Yo-Yo.ai Gitbook Repository
  • Prediction: Accuracy
    • Overview
    • Technical
  • Model Output
  • Case Study
  • Conclusion
  • Prediction: Alpha
    • Overview
  • Methodology
  • Feature Engineering
  • Model Ensemble Implementation
  • Risk Management
  • Model Testing and Backtesting
    • Testing Methodology
    • Backtesting Framework
    • Hybrid Strategy Implementation
    • Results
    • Key Findings
    • APPENDIX A:
  • Vaults
    • How To
  • Product Road Map
  • FAQ
  • YOYO Token
    • Tokenomics
      • The YOYO Token
      • Token Utility
      • Token Metrics
  • Branding
    • Logos
  • Media
  • Legal Documents
    • Terms and Conditions
    • Privacy Policy
    • Disclaimer
  • Official Links
    • Official Links
Powered by GitBook
On this page
  • 1.1 Prediction Framework
  • 1.2 AI Strategy

Methodology

PreviousOverviewNextFeature Engineering

Last updated 2 months ago

1.1 Prediction Framework

1.2 AI Strategy

The AI strategy, implemented on a unified account with capital C, dynamically adjusts positions based on the consensus derived from multiple ML models. This approach maintains a directional bias, ensuring that the overall strategy remains aligned with the prevailing market trend.

1.2.1 Key Characteristics & Advantages

  • Directional Persistence: The strategy leverages the power of trend-following by maintaining a consistent directional bias. This means that it aims to capitalize on sustained market movements, whether bullish or bearish.

  • Continuous Position Sizing: Position sizes are not static but are continuously adjusted in response to the evolving market conditions and the consensus of the ML models. This allows for greater flexibility and responsiveness to market dynamics.

  • Transaction Cost Optimization: The strategy incorporates mechanisms to minimize transaction costs, which can have a significant impact on overall profitability.