29 January 2024 in IEEE Access
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.
2 July 2024 in Expert Systems
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.
12 February 2025 in Business Strategy and the Environment
This research examines how climate claims by companies from the United Kingdom have changed over the years, especially when they became certain about the mandate of the Taskforce on Climate‐related Financial Disclosure (TCFD). We use text data from FTSE 100 companies for eight consecutive years, starting from 2016, and apply the robust ClimateBERT algorithm to analyse company statements related to climate claims, where they claim how they take care of climate in their business operations. Our findings show that the total number of corporate climate claims made has substantially increased since 2016, resulting in an overall improvement in corporate environmental claims till 2023. This coincides with the official announcement of the TCFD mandate.Our analyses also indicate that the proportion of claims in each report has increased over the years despite economic uncertainties. Additionally, the study findings reveal that even industries with minimal or negligible climate claims can still be associated with carbon‐intensive activities. The complementary features of the legitimacy and stakeholder theories support our findings. By applying ClimateBERT, our research mitigates existing data challenges, yielding an efficient framework for analysing text through a robust natural language processing model. Our findings will assist policymakers in identifying necessary modifications to corporate climate disclosure and will help assess the impact of the Taskforce intervention on climate‐related financial disclosure.
3 April 2026 in International Journal of Electronic Commerce