Examinando por Autor "Wu, Jian"

Mostrando 1 - 3 de 3
  • Miniatura
    Ítem
    Autoencoder-enhanced clustering: a dimensionality reduction approach to financial time series
    (Institute of Electrical and Electronics Engineers Inc., 2024-02-05) González Cortés, Daniel Alejandro; Onieva Caracuel, Enrique; Pastor López, Iker; Trinchera, Laura; Wu, Jian
    While Machine Learning significantly boosts the performance of predictive models, its efficacy varies across different data dimensions. It is essential to cluster time series data of similar characteristics, particularly in the financial sector. However, clustering financial time series data poses considerable challenges due to the market's inherent complexity and multidimensionality. To address these issues, our study introduces a novel clustering framework that leverages autoencoders for a compressed yet informative representation of financial time series. We rigorously evaluate our approach through multiple dimensionality reduction and clustering algorithms, applying it to key financial indices, including IBEX-35, CAC-40, DAX-30, S&P 500, and FTSE 100. Our findings consistently demonstrate that incorporating autoencoders significantly enhances the granularity and quality of clustering, effectively isolating distinct categories of financial time series. Our findings carry significant ramifications for the financial industry. By refining clustering methodologies, we set the stage for increasingly accurate financial predictive models, offering valuable insights for optimizing investment strategies and enhancing risk management.
  • Miniatura
    Ítem
    Portfolio construction using explainable reinforcement learning
    (John Wiley and Sons Inc, 2024-11) Cortés González, Daniel; Onieva Caracuel, Enrique; Pastor López, Iker; Trinchera, Laura; Wu, Jian
    While machine learning's role in financial trading has advanced considerably, algorithmic transparency and explainability challenges still exist. This research enriches prior studies focused on high-frequency financial data prediction by introducing an explainable reinforcement learning model for portfolio management. This model transcends basic asset prediction, formulating concrete, actionable trading strategies. The methodology is applied in a custom trading environment mimicking the CAC-40 index's financial conditions, allowing the model to adapt dynamically to market changes based on iterative learning from historical data. Empirical findings reveal that the model outperforms an equally weighted portfolio in out-of-sample tests. The study offers a dual contribution: it elevates algorithmic planning while significantly boosting transparency and interpretability in financial machine learning. This approach tackles the enduring ‘black-box’ issue and provides a holistic, transparent framework for managing investment portfolios.
  • No hay miniatura disponible
    Ítem
    Predicting cryptocurrency prices during economic uncertainty with explainable Artificial Intelligence
    (Routledge, 2026) González Cortés, Daniel Alejandro ; Nandy, Monomita; Suman Lodh,; Senyo, P.K.; Wu, Jian; Onieva Caracuel, Enrique
    Predicting cryptocurrency prices is challenging due to their high volatility. This challenge is more pronounced during economic uncertainty, such as the 2008 financial crisis and the COVID-19 pandemic. While machine learning models can help in the prediction of cryptocurrency prices, their underlying conditions influencing the outcomes are sometimes unknown, and there is a lack of consensus on appropriate techniques to use for technical prediction and circumstances under which they may be suitable. In this paper, we apply an existing explainable artificial intelligence (XAI) framework, specifically SHAP, to identify suitable analytical techniques and the optimized set of parameters for technical trading prediction based on the two most valuable cryptocurrencies, Bitcoin and Ethereum. Rather than developing a new model, our contribution lies in systematically applying XAI techniques to uncover variable importance and model behavior in volatile market conditions. The results show that our explainable AI model is capable of efficiently forecasting closing, high, and low prices from previous days during economic uncertainties. Through our model and findings, we contribute critical insights to research and practice, especially in overcoming the challenges of the “black box” nature of machine learning models. Moreover, practitioners such as investors and regulators can utilize our model to efficiently capture changes in different cryptocurrencies’ price trends toward improved decision-making during economic uncertainty.