STAGE framework: A stock dynamic anomaly detection and trend prediction model based on graph attention network and sparse spatiotemporal convolutional network.

As the financial market becomes increasingly complex, stock prediction and anomaly data detection have emerged as crucial tasks in financial risk management. However, existing methods exhibit significant limitations in handling the intricate relationships between stocks and addressing anomalous data...

Full description

Saved in:
Bibliographic Details
Main Authors: Ming Shi, Roznim Mohamad Rasli, Shir Li Wang
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-01-01
Series:PLoS ONE
Online Access:https://doi.org/10.1371/journal.pone.0318939
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849714888633483264
author Ming Shi
Roznim Mohamad Rasli
Shir Li Wang
author_facet Ming Shi
Roznim Mohamad Rasli
Shir Li Wang
author_sort Ming Shi
collection DOAJ
description As the financial market becomes increasingly complex, stock prediction and anomaly data detection have emerged as crucial tasks in financial risk management. However, existing methods exhibit significant limitations in handling the intricate relationships between stocks and addressing anomalous data. This paper proposes the STAGE framework, which integrates the Graph Attention Network (GAT), Variational Autoencoder (VAE), and Sparse Spatiotemporal Convolutional Network (STCN), to enhance the accuracy of stock prediction and the robustness of anomaly data detection. Experimental results show that the complete STAGE framework achieved an accuracy of 85% after 20 training epochs, which is 10% to 20% higher than models with key algorithms removed. In the anomaly detection task, the STAGE framework further improved the accuracy to 95%, demonstrating fast convergence and stability. This framework offers an innovative solution for stock prediction, adapting to the complex dynamics of real-world markets.
format Article
id doaj-art-9852c242eda54d2bafd1db2dd44d7cb4
institution DOAJ
issn 1932-6203
language English
publishDate 2025-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj-art-9852c242eda54d2bafd1db2dd44d7cb42025-08-20T03:13:33ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01203e031893910.1371/journal.pone.0318939STAGE framework: A stock dynamic anomaly detection and trend prediction model based on graph attention network and sparse spatiotemporal convolutional network.Ming ShiRoznim Mohamad RasliShir Li WangAs the financial market becomes increasingly complex, stock prediction and anomaly data detection have emerged as crucial tasks in financial risk management. However, existing methods exhibit significant limitations in handling the intricate relationships between stocks and addressing anomalous data. This paper proposes the STAGE framework, which integrates the Graph Attention Network (GAT), Variational Autoencoder (VAE), and Sparse Spatiotemporal Convolutional Network (STCN), to enhance the accuracy of stock prediction and the robustness of anomaly data detection. Experimental results show that the complete STAGE framework achieved an accuracy of 85% after 20 training epochs, which is 10% to 20% higher than models with key algorithms removed. In the anomaly detection task, the STAGE framework further improved the accuracy to 95%, demonstrating fast convergence and stability. This framework offers an innovative solution for stock prediction, adapting to the complex dynamics of real-world markets.https://doi.org/10.1371/journal.pone.0318939
spellingShingle Ming Shi
Roznim Mohamad Rasli
Shir Li Wang
STAGE framework: A stock dynamic anomaly detection and trend prediction model based on graph attention network and sparse spatiotemporal convolutional network.
PLoS ONE
title STAGE framework: A stock dynamic anomaly detection and trend prediction model based on graph attention network and sparse spatiotemporal convolutional network.
title_full STAGE framework: A stock dynamic anomaly detection and trend prediction model based on graph attention network and sparse spatiotemporal convolutional network.
title_fullStr STAGE framework: A stock dynamic anomaly detection and trend prediction model based on graph attention network and sparse spatiotemporal convolutional network.
title_full_unstemmed STAGE framework: A stock dynamic anomaly detection and trend prediction model based on graph attention network and sparse spatiotemporal convolutional network.
title_short STAGE framework: A stock dynamic anomaly detection and trend prediction model based on graph attention network and sparse spatiotemporal convolutional network.
title_sort stage framework a stock dynamic anomaly detection and trend prediction model based on graph attention network and sparse spatiotemporal convolutional network
url https://doi.org/10.1371/journal.pone.0318939
work_keys_str_mv AT mingshi stageframeworkastockdynamicanomalydetectionandtrendpredictionmodelbasedongraphattentionnetworkandsparsespatiotemporalconvolutionalnetwork
AT roznimmohamadrasli stageframeworkastockdynamicanomalydetectionandtrendpredictionmodelbasedongraphattentionnetworkandsparsespatiotemporalconvolutionalnetwork
AT shirliwang stageframeworkastockdynamicanomalydetectionandtrendpredictionmodelbasedongraphattentionnetworkandsparsespatiotemporalconvolutionalnetwork