SPPMFN: Efficient Multimodal Financial Time-Series Prediction Network With Self-Supervised Learning
Financial time series exhibit high volatility and non-linearity, making analysis particularly challenging. Traditional statistical methods like ARIMA and GARCH struggle with non-linear data. At the same time, despite capturing complex price dynamics, machine learning and deep learning approaches oft...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11104243/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849239054395113472 |
|---|---|
| author | Ningxin Li Gang Chao Jianke Zou Gaozhe Jiang |
| author_facet | Ningxin Li Gang Chao Jianke Zou Gaozhe Jiang |
| author_sort | Ningxin Li |
| collection | DOAJ |
| description | Financial time series exhibit high volatility and non-linearity, making analysis particularly challenging. Traditional statistical methods like ARIMA and GARCH struggle with non-linear data. At the same time, despite capturing complex price dynamics, machine learning and deep learning approaches often face the risk of overfitting. Additionally, the limited parameters of one-dimensional financial time series signals constrain feature representation. To tackle these challenges, we propose an efficient multimodal financial time series prediction network utilizing self-supervised learning based on the custom-designed SPPMFN network for stock trend forecasting. Our approach introduces a novel signal transformation strategy to extract richer multi-scale feature representations from financial time series signals. Specifically, we convert one-dimensional stock price time series into two-dimensional image sequences across varying time intervals using Gramian Angular Fields. These transformed data modalities are then simultaneously processed by the SPPMFN, enabling the extraction of features from multiple dimensions. Furthermore, we introduce a self-supervised learning framework that enhances the model’s ability to capture intrinsic relationships within the data, allowing it to detect underlying patterns while effectively mitigating overfitting. Experimental results on the CSI300E, CSI100E, and S&P500 datasets demonstrate the effectiveness of our method, showing superior performance in accurately predicting high-yield stocks and significantly outperforming industry benchmarks. Specifically, our model improves accuracy by over 3% and achieves a cumulative return of 21.62% on the CSI300E dataset. On the S&P500 dataset, accuracy increased by over 1.4% and cumulative return reached 18.66%. These findings highlight the robustness and practical applicability of our method across diverse financial markets. |
| format | Article |
| id | doaj-art-9e40c408920541a09d9a854bcc229801 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-9e40c408920541a09d9a854bcc2298012025-08-20T04:01:15ZengIEEEIEEE Access2169-35362025-01-011313488313489710.1109/ACCESS.2025.359385111104243SPPMFN: Efficient Multimodal Financial Time-Series Prediction Network With Self-Supervised LearningNingxin Li0Gang Chao1https://orcid.org/0009-0002-4576-1949Jianke Zou2Gaozhe Jiang3https://orcid.org/0009-0006-8015-9223Fu Foundation School of Engineering and Applied Science, Columbia University, New York, NY, USAFaculty of Business and Management, Beijing Normal University-Hong Kong Baptist University United International College, Zhuhai, ChinaDepartment of Management, HSBC Business School, Peking University, Beijing, ChinaInstitute of Operations Research and Analytics, National University of Singapore, Queenstown, SingaporeFinancial time series exhibit high volatility and non-linearity, making analysis particularly challenging. Traditional statistical methods like ARIMA and GARCH struggle with non-linear data. At the same time, despite capturing complex price dynamics, machine learning and deep learning approaches often face the risk of overfitting. Additionally, the limited parameters of one-dimensional financial time series signals constrain feature representation. To tackle these challenges, we propose an efficient multimodal financial time series prediction network utilizing self-supervised learning based on the custom-designed SPPMFN network for stock trend forecasting. Our approach introduces a novel signal transformation strategy to extract richer multi-scale feature representations from financial time series signals. Specifically, we convert one-dimensional stock price time series into two-dimensional image sequences across varying time intervals using Gramian Angular Fields. These transformed data modalities are then simultaneously processed by the SPPMFN, enabling the extraction of features from multiple dimensions. Furthermore, we introduce a self-supervised learning framework that enhances the model’s ability to capture intrinsic relationships within the data, allowing it to detect underlying patterns while effectively mitigating overfitting. Experimental results on the CSI300E, CSI100E, and S&P500 datasets demonstrate the effectiveness of our method, showing superior performance in accurately predicting high-yield stocks and significantly outperforming industry benchmarks. Specifically, our model improves accuracy by over 3% and achieves a cumulative return of 21.62% on the CSI300E dataset. On the S&P500 dataset, accuracy increased by over 1.4% and cumulative return reached 18.66%. These findings highlight the robustness and practical applicability of our method across diverse financial markets.https://ieeexplore.ieee.org/document/11104243/Multimodal stock trend predictionsseries-to-imageneural networkself-supervised learning |
| spellingShingle | Ningxin Li Gang Chao Jianke Zou Gaozhe Jiang SPPMFN: Efficient Multimodal Financial Time-Series Prediction Network With Self-Supervised Learning IEEE Access Multimodal stock trend predictions series-to-image neural network self-supervised learning |
| title | SPPMFN: Efficient Multimodal Financial Time-Series Prediction Network With Self-Supervised Learning |
| title_full | SPPMFN: Efficient Multimodal Financial Time-Series Prediction Network With Self-Supervised Learning |
| title_fullStr | SPPMFN: Efficient Multimodal Financial Time-Series Prediction Network With Self-Supervised Learning |
| title_full_unstemmed | SPPMFN: Efficient Multimodal Financial Time-Series Prediction Network With Self-Supervised Learning |
| title_short | SPPMFN: Efficient Multimodal Financial Time-Series Prediction Network With Self-Supervised Learning |
| title_sort | sppmfn efficient multimodal financial time series prediction network with self supervised learning |
| topic | Multimodal stock trend predictions series-to-image neural network self-supervised learning |
| url | https://ieeexplore.ieee.org/document/11104243/ |
| work_keys_str_mv | AT ningxinli sppmfnefficientmultimodalfinancialtimeseriespredictionnetworkwithselfsupervisedlearning AT gangchao sppmfnefficientmultimodalfinancialtimeseriespredictionnetworkwithselfsupervisedlearning AT jiankezou sppmfnefficientmultimodalfinancialtimeseriespredictionnetworkwithselfsupervisedlearning AT gaozhejiang sppmfnefficientmultimodalfinancialtimeseriespredictionnetworkwithselfsupervisedlearning |