Trading digital assets like Bitcoin, Ethereum, and altcoins.
An explainable multi-agent system using fine-tuned GPT-4o models for crypto portfolio management. Specialized agents analyze news, factors, and charts, collaborate on decisions, and execute trades—outperforming benchmarks in returns, accuracy, and interpretability.
ChatGPT can classify crypto tweets into bullish or bearish and predict Bitcoin returns better than BERT or VADER. Its sentiment signals show strong predictive power for daily returns.
Bitcoin’s extreme volatility comes with an upside: low correlation with other assets. This paper shows that even a small allocation to Bitcoin can significantly boost the Sharpe ratio of a diversified portfolio.
This paper shows that Initial Coin Offerings (ICOs) can help startups grow by giving them access to funding and liquidity. Projects that make good disclosures, use tokens with real utility, and get listed on exchanges are more likely to survive and hire more employees later.
This paper shows that cryptocurrency prices can be much higher in some countries than others. These price gaps happen because of restrictions on moving money between countries, not because of problems with the cryptocurrencies themselves.
Crypto returns aren’t driven by mining costs or fiat correlations. Instead, they respond to network adoption, momentum, and investor attention. This paper identifies the key factors that predict crypto returns and provides a foundation for data-driven crypto investing.