Deep Learning-Based Hybrid Model with Multi-Head Attention for Multi-Horizon Stock Price Prediction

View Full Paper

Ask AI

Bookmark

The model achieves superior predictive performance for stock prices over varying horizons of 1, 3, and 7 days.
Using SHAP analysis verifies the reliability of the selected features, enhancing the model's accuracy.
The proposed architecture incorporates a temporal convolutional network and gated recurrent unit to improve predictions.
Lower error rates and higher accuracy are consistently observed when compared to benchmark models like LSTM and GRU.

Discussion

No takes yet. Share an insight, caveat, or question.

A deep learning model improves stock price predictions, indicating better accuracy with feature selection.

Ghosh et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark