arXiv:2606.15701v1 Announce Type: new Abstract: Transformers have shown remarkable success in sequence modeling, yet their direct application to financial time series remains challenging due to noisy signals, short-memory dynamics, and distributional shifts. This paper proposes a modified Transformer architecture for one-step stock index forecasting, combined with advanced learning-rate scheduling and a novel Shifted Data Augmentation (SDA) technique. We evaluate the proposed framework on two benchmark stock index datasets, VN30 and S&P 500. Experimental results demonstrate that cosine anneali
The paper leverages recent advancements in Transformer architectures and data augmentation techniques to address existing challenges in applying AI to volatile financial time series data.
Improved stock index forecasting methods can enhance decision-making for institutional investors and highlight the expanding capabilities of AI in complex domains.
The proposed modified Transformer and Shifted Data Augmentation technique offer a more robust and accurate approach to short-term financial prediction.
- · Quantitative hedge funds
- · Asset managers
- · Fintech companies
- · Traditional qualitative analysts
- · Inefficient trading strategies
More accurate stock index predictions become possible, leading to enhanced trading and investment strategies.
Increased adoption of advanced AI models in finance could further automate trading and reduce reliance on human discretion.
The competitive landscape of financial markets may intensify as sophisticated AI tools become more widespread, potentially centralizing power among those with access to superior models and computational resources.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG