A Deep Learning Framework for High-Frequency Signal Forecasting Based on Graph and Temporal-Macro Fusion
With the increase in trading frequency and the growing complexity of data structures, traditional quantitative strategies have gradually encountered bottlenecks in modeling capacity, real-time responsiveness, and multi-dimensional information integration. To address these limitations, a high-frequen...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-04-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/9/4605 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849312751146500096 |
|---|---|
| author | Xijue Zhang Liman Zhang Siyang He Tianyue Li Yinke Huang Yaqi Jiang Haoxiang Yang Chunli Lv |
| author_facet | Xijue Zhang Liman Zhang Siyang He Tianyue Li Yinke Huang Yaqi Jiang Haoxiang Yang Chunli Lv |
| author_sort | Xijue Zhang |
| collection | DOAJ |
| description | With the increase in trading frequency and the growing complexity of data structures, traditional quantitative strategies have gradually encountered bottlenecks in modeling capacity, real-time responsiveness, and multi-dimensional information integration. To address these limitations, a high-frequency signal generation framework is proposed, which integrates graph neural networks, cross-scale Transformer architectures, and macro factor modeling. This framework enables unified modeling of structural dependencies, temporal fluctuations, and macroeconomic disturbances. In predictive validation experiments, the framework achieved a precision of 92.4%, a recall of 91.6%, and an F1-score of 92.0% on classification tasks. For regression tasks, the mean squared error (MSE) and mean absolute error (MAE) were reduced to <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>1.76</mn><mo>×</mo><msup><mn>10</mn><mrow><mo>−</mo><mn>4</mn></mrow></msup></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.96</mn><mo>×</mo><msup><mn>10</mn><mrow><mo>−</mo><mn>2</mn></mrow></msup></mrow></semantics></math></inline-formula>, respectively. These results significantly outperformed several mainstream models, including LSTM, FinBERT, and StockGCN, demonstrating superior stability and practical applicability. |
| format | Article |
| id | doaj-art-cbbafc442d7d4aebbeed886bdac40b18 |
| institution | Kabale University |
| issn | 2076-3417 |
| language | English |
| publishDate | 2025-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-cbbafc442d7d4aebbeed886bdac40b182025-08-20T03:52:57ZengMDPI AGApplied Sciences2076-34172025-04-01159460510.3390/app15094605A Deep Learning Framework for High-Frequency Signal Forecasting Based on Graph and Temporal-Macro FusionXijue Zhang0Liman Zhang1Siyang He2Tianyue Li3Yinke Huang4Yaqi Jiang5Haoxiang Yang6Chunli Lv7China Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaChina Agricultural University, Beijing 100083, ChinaWith the increase in trading frequency and the growing complexity of data structures, traditional quantitative strategies have gradually encountered bottlenecks in modeling capacity, real-time responsiveness, and multi-dimensional information integration. To address these limitations, a high-frequency signal generation framework is proposed, which integrates graph neural networks, cross-scale Transformer architectures, and macro factor modeling. This framework enables unified modeling of structural dependencies, temporal fluctuations, and macroeconomic disturbances. In predictive validation experiments, the framework achieved a precision of 92.4%, a recall of 91.6%, and an F1-score of 92.0% on classification tasks. For regression tasks, the mean squared error (MSE) and mean absolute error (MAE) were reduced to <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>1.76</mn><mo>×</mo><msup><mn>10</mn><mrow><mo>−</mo><mn>4</mn></mrow></msup></mrow></semantics></math></inline-formula> and <inline-formula><math xmlns="http://www.w3.org/1998/Math/MathML" display="inline"><semantics><mrow><mn>0.96</mn><mo>×</mo><msup><mn>10</mn><mrow><mo>−</mo><mn>2</mn></mrow></msup></mrow></semantics></math></inline-formula>, respectively. These results significantly outperformed several mainstream models, including LSTM, FinBERT, and StockGCN, demonstrating superior stability and practical applicability.https://www.mdpi.com/2076-3417/15/9/4605deep learning in quantitative schemesignal forecastinggraph neural networksattention mechanism |
| spellingShingle | Xijue Zhang Liman Zhang Siyang He Tianyue Li Yinke Huang Yaqi Jiang Haoxiang Yang Chunli Lv A Deep Learning Framework for High-Frequency Signal Forecasting Based on Graph and Temporal-Macro Fusion Applied Sciences deep learning in quantitative scheme signal forecasting graph neural networks attention mechanism |
| title | A Deep Learning Framework for High-Frequency Signal Forecasting Based on Graph and Temporal-Macro Fusion |
| title_full | A Deep Learning Framework for High-Frequency Signal Forecasting Based on Graph and Temporal-Macro Fusion |
| title_fullStr | A Deep Learning Framework for High-Frequency Signal Forecasting Based on Graph and Temporal-Macro Fusion |
| title_full_unstemmed | A Deep Learning Framework for High-Frequency Signal Forecasting Based on Graph and Temporal-Macro Fusion |
| title_short | A Deep Learning Framework for High-Frequency Signal Forecasting Based on Graph and Temporal-Macro Fusion |
| title_sort | deep learning framework for high frequency signal forecasting based on graph and temporal macro fusion |
| topic | deep learning in quantitative scheme signal forecasting graph neural networks attention mechanism |
| url | https://www.mdpi.com/2076-3417/15/9/4605 |
| work_keys_str_mv | AT xijuezhang adeeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT limanzhang adeeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT siyanghe adeeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT tianyueli adeeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT yinkehuang adeeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT yaqijiang adeeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT haoxiangyang adeeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT chunlilv adeeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT xijuezhang deeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT limanzhang deeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT siyanghe deeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT tianyueli deeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT yinkehuang deeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT yaqijiang deeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT haoxiangyang deeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion AT chunlilv deeplearningframeworkforhighfrequencysignalforecastingbasedongraphandtemporalmacrofusion |