The field of financial forecasting and analysis is witnessing significant developments, with a growing emphasis on hybrid models that combine traditional techniques with machine learning and deep learning approaches. Researchers are exploring the potential of these models to improve the accuracy of financial predictions, including stock prices, unemployment rates, and asset pricing. The integration of natural language processing and sentiment analysis is also becoming increasingly important, as it allows for the incorporation of qualitative factors into financial models. Furthermore, the development of new libraries and frameworks, such as fastabx and LLM4FTS, is facilitating the efficient computation of financial metrics and the enhancement of large language models for financial time series prediction. Noteworthy papers include:
- LLM4FTS, which proposes a novel framework for enhancing large language models for financial time series prediction,
- Representation Learning of Limit Order Book, which conducts a comprehensive study of limit order book representation learning,
- Dynamic Asset Pricing, which integrates FinBERT-based sentiment quantification with the Fama-French five-factor model.