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. Prediction: Accuracy

Overview

DLSTM Model: An Entropy-Based Approach to Prediction Accuracy

Yo-Yo’s research team has developed a proprietary AI-driven price prediction model with a breakthrough 75% average accuracy rate in forecasting, surpassing the 63% historical industry benchmark. As a result, the Yo-Yo AI model is being trialed by a $300 million programmatic hedge fund. With a demonstrated 75% accuracy rate, Yo-Yo’s AI-driven prediction model utilizes a sophisticated neural network architecture designed to forecast mid-price movements across future time intervals of up to 3 hours. At scale, Yo-Yo can predict the future state of multiple assets, or even entire stock markets, making it extensible across crypto, equities, commodities, forex and derivatives.

PreviousYo-Yo.ai Gitbook RepositoryNextTechnical

Last updated 2 months ago