The next meeting of the seminar of the Information Modeling Department will be held on

May 26, 2023, at 2:00 pm, in Room 278 of IMI.

A talk on

Optimal Algorithmic Trading by Means of AI

will be delivered by Senior Assist. Prof. Dr. Slavi Georgiev.

Abstract. This study delves into an intriguing investigation of the capabilities and potential of ChatGPT-4, an advanced language model, in enhancing algorithmic trading strategies for cryptocurrencies. Specifically, the focus lies on the application of automated bots in this domain. Recognizing the dynamic nature of the cryptocurrency market, the authors aim to leverage the unique features of ChatGPT-4 to develop a sophisticated trading bot that can navigate this complex landscape effectively.

The first crucial step in this study involves comprehensively understanding the current market structure of cryptocurrencies. By analyzing various factors such as liquidity, volatility, and market depth, the researchers gain valuable insights into the nuances and intricacies of this ever-evolving market. These insights lay the foundation for developing a robust trading strategy.

In addition to market structure analysis, the implementation of appropriate risk management techniques is a key aspect of this research. Given the inherent volatility and uncertainty associated with cryptocurrencies, effective risk management is essential to safeguarding the trading bot’s performance and ensuring its longevity. The authors explore various risk management approaches, such as portfolio diversification, stop-loss orders, and position sizing, to mitigate potential losses and enhance overall risk-adjusted returns.

To validate the effectiveness of the proposed ChatGPT-4-based trading bot, the authors employ rigorous backtesting simulations using tick data. By leveraging historical market data and applying the trading strategy developed with ChatGPT-4, the performance of the trading bot is thoroughly evaluated under realistic market conditions. This process helps to assess the bot’s ability to adapt to different market scenarios, capture profitable trading opportunities, and maintain consistent performance over time.

By leveraging the advanced natural language processing capabilities of ChatGPT-4, this study opens up new ways for improving algorithmic trading in the cryptocurrency market. The integration of language understanding and sophisticated trading algorithms can potentially provide traders with a powerful tool for making informed decisions and executing trades in a more efficient and profitable manner. Ultimately, the findings of this research contribute to the advancement of algorithmic trading techniques in the realm of cryptocurrencies, offering new insights and opportunities for market participants.

Everybody is invited.