Journal of Risk and Financial Management

19 pages, 3417 KB

Open AccessArticle

Does CNN-Based Feature Extraction Improve High-Frequency Return Prediction? Evidence from the CSI 300 Index

by Fan Zhang and Haobing Wang

J. Risk Financial Manag. 2026, 19(5), 371; https://doi.org/10.3390/jrfm19050371 - 20 May 2026

This study investigates whether CNN-based front-end feature extraction improves the predictive performance of deep learning models applied to 1 min intraday CSI 300 index data. Three baseline sequence models, LSTM, GRU, and TCN, are compared against their CNN hybrid and dual-branch fusion variants [...] Read more.

This study investigates whether CNN-based front-end feature extraction improves the predictive performance of deep learning models applied to 1 min intraday CSI 300 index data. Three baseline sequence models, LSTM, GRU, and TCN, are compared against their CNN hybrid and dual-branch fusion variants across five input window sizes, with all comparisons using identical back-end configurations. A total of 45 model configurations are trained and evaluated across 20 independent runs, with performance assessed on four metrics (MAE, RMSE, Directional Accuracy, and Information Coefficient) and statistical significance evaluated by paired t-tests. After standardisation, adding a CNN front-end does not consistently improve performance over the raw baseline and reduces IC for LSTM- and GRU-based models in many cases (e.g., IC of 0.0187 vs. 0.1031 for CNN-LSTM vs. LSTM at

W = 1

), suggesting that standardised recurrent models can extract useful patterns directly from the raw sequence without CNN preprocessing. The dual-branch fusion architecture, which retains both the raw and CNN-compressed sequence branches, consistently outperforms the pure CNN hybrid on MAE, RMSE, and IC for LSTM- and GRU-based models (e.g., LSTMDualBranchFusion achieves statistically significant MAE reductions over CNN-LSTM at

W = 1

,

W = 2

,

W = 4

, and

W = 5

), indicating that the raw sequence carries complementary predictive information that the CNN front-end discards. TCN-based models produce near-zero or negative IC values regardless of architecture variant, suggesting a possible limitation of dilated convolutional architectures for return rank-ordering on this dataset and sample period. These findings are consistent across all five window sizes examined. Full article

(This article belongs to the Special Issue Quantitative Finance in the Era of Big Data and AI)

► Show Figures

Figure 1

18 pages, 276 KB

Open AccessArticle

Intervening Influence of Financial Development on the Relationship Between Sustainability Practices and Sustainable Development of the Sub-Saharan African Countries

by James C. N. Mbugua, Ibrahim Tirimba Ondabu and Fred Ochogo Sporta

J. Risk Financial Manag. 2026, 19(5), 370; https://doi.org/10.3390/jrfm19050370 - 20 May 2026

Journal Description

Journal of Risk and Financial Management

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI