Boosting the Accuracy of Stock Market Prediction via Multi-Layer Hybrid MTL Structure

TL;DR

This paper introduces a multi-layer hybrid multi-task learning framework combining Transformer, BiGRU, and KAN to improve stock market prediction accuracy, effectively handling complex financial data characteristics.

Contribution

It presents a novel hybrid MTL structure integrating multiple neural components for enhanced stock prediction performance, outperforming existing models.

Findings

Achieved MAE as low as 1.078

Achieved MAPE as low as 0.012

Achieved R^2 as high as 0.98

Abstract

Accurate stock market prediction provides great opportunities for informed decision-making, yet existing methods struggle with financial data's non-linear, high-dimensional, and volatile characteristics. Advanced predictive models are needed to effectively address these complexities. This paper proposes a novel multi-layer hybrid multi-task learning (MTL) framework aimed at achieving more efficient stock market predictions. It involves a Transformer encoder to extract complex correspondences between various input features, a Bidirectional Gated Recurrent Unit (BiGRU) to capture long-term temporal relationships, and a Kolmogorov-Arnold Network (KAN) to enhance the learning process. Experimental evaluations indicate that the proposed learning structure achieves great performance, with an MAE as low as 1.078, a MAPE as low as 0.012, and an R^2 as high as 0.98, when compared with other competitive networks.