[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-152924-105":59,"doc-detail-152924-en":130},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","sgat-sp-sparse-graph-attention-network-for-stock-prediction-research-article","SGAT-SP: Sparse Graph Attention Network for Stock Prediction - Research Article","","Industry experts and academic scholars focus on stock market volatility, and earlier work often relies on single inputs such as closing prices, opening prices, or news. Interest has increased in multimodal graph neural networks that jointly model related stocks using graph structures. The proposed SGAT-SP builds on multimodal graph attention by addressing ML-GAT’s noisy intercompany edges and inefficient per-layer attention coefficients. SGAT-SP uses shared attention coefficients and a binary edge mask to suppress irrelevant links during node aggregation. Results show average accuracy of 0.83, slightly higher than ML-GAT (0.827), and an inference time reduction of 85%.",{"@graph":69,"@context":122},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/sgat-sp-sparse-graph-attention-network-for-stock-prediction-research-article/152924/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/sgat-sp-sparse-graph-attention-network-for-stock-prediction-research-article/152924.png","ImageObject",300,407,{"name":92,"@type":93},"Miles","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-09-21","2026-08-27",true,{"@type":102,"interactionType":103,"userInteractionCount":24},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"What problem does SGAT-SP address in stock prediction using graph attention networks?","Question",{"text":112,"@type":113},"It addresses noisy and irrelevant intercompany relationships that can arise in prior graph attention approaches, as well as inefficiencies caused by computing attention coefficients separately across layers.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"How does SGAT-SP reduce noisy edges in the stock graph?",{"text":117,"@type":113},"It assigns a binary mask to edges to decide whether an edge is used for node aggregation, suppressing noisy connections during training and inference.",{"name":119,"@type":110,"acceptedAnswer":120},"What performance improvements does SGAT-SP report compared with ML-GAT?",{"text":121,"@type":113},"SGAT-SP achieves an average accuracy of 0.83 versus ML-GAT’s 0.827, and it reduces inference time by 85%, lowering computational expenses.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},152924,1787868330,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":24,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":129,"read_time":41},13056703019404,"https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0","Journal of Forecasting  \nRESEARCH ARTICLE  \nSGAT-SP: Sparse Graph Attention Network for Stock Prediction  \nGautam Gupta | Priyanshi Goel | Divyanshi Verma | Amarjit Malhotra  Netaji Subhas University of Technology, Delhi, India  \nCorrespondence: Amarjit Malhotra ([amarjit.malhotra@nsut.ac.in](amarjit.malhotra@nsut.ac.in))  \nReceived: 6 March 2024 | Revised: 5 March 2026 | Accepted: 23 April 2026  \nABSTRACT  \nIndustry experts as well as academic scholars have directed substantial attention toward researching stock market volatility. Most of the past research has focused on using singular features such as closing prices, opening prices, or news stories to predict stock movement. In recent times, there has been a growing interest in using multimodal graph neural networks which can analyze a variety of features. Most of these methods focus on node aggregation to extract relevant features from related stocks to further improve model accuracy. The current state-of-the-art model—ML-GAT: Multilevel Graph Attention Model—constructsa graph network between the stocks using Wikidata relations. It uses multiple layers of graph attention to aggregate features such as historical price features and current news. However, the high number of intercompany relationships in ML-GAT may include irrelevant and noisy edges. It also uses individual attention coefficients for each layer, leading to inefficient utilization of computational resources. To overcome these challenges, a Sparse Graph Attention Network for Stock Prediction (SGAT-SP) is proposed in this paper. SGAT-SP uses a single set of attention coefficients to reduce training and inference time. It assigns a binary mask to every edge which represents whether they will be used for node aggregation to reduce noisy edges. The proposed approach achieves an average accuracy score of 0.83, a slight improvement over ML-GAT, which has an accuracy score of 0.827. Additionally, it significantly reduces inference time by 85%, resulting in faster results and decreased computational expenses.  \n1 | Introduction  \nThe importance of stock price data in real-life scenarios is extremely crucial for experts, researchers, and scholars to grasp to understand the economic conditions of a country. The objective of stock price prediction is to foresee the variation in stock prices to help stockholders make decisions about buying and selling a particular stock. Through the use of machine learning algorithms and computations on statistical data, it is possible to ascertain the future valuation of a stock (Kumari et al. 2021) . Given the dynamic nature of stock markets, various approaches can be adopted to predict prices, including fundamental, technical, and quantitative analysis (Nti et al. 2019) . Fundamental analysis involves comprehending a company's financial strength, valuation, and other relevant factors to estimate the approximate stock price. Technical studies focus on historical price data to  \nidentify market trends, which is particularly useful for shortterm predictions. Quantitative analysis focuses on historical price data as well as real-time company relations data to make predictions to provide valuable insights into market trends and investor sentiments, offering a greater understanding of market dynamics than just numerical metrics.  \nFurther, the accuracy of such predictions is dependent on data sources such as historical prices and news articles. Most of the studies have predominantly focused only on individual pieces of information, ignoring the closely related factors. Since these parameters are interrelated, each significantly affects the other and hence must be considered together to enhance the prediction accuracy. The combination of multiple data sources allows the model to gain a richer understanding of the stock market trends.  \n© 2026 John Wiley & Sons Ltd.  \n2942 Journal of Forecasting, 2026; 45:2942–2953  \n[https://doi.org/10.1002/for.70169](https://doi.org/10.1002/for.70169)  \nThe noi","cbCailhT2FoehSom","https://ap.wps.com/l/cbCailhT2FoehSom","pdf",1261468,12,"English","# Introduction\n## Stock prediction goals and analysis types\n## Data sources and interrelated factors\n## Motivation for graph-based attention methods","[{\"question\":\"What problem does SGAT-SP address in stock prediction using graph attention networks?\",\"answer\":\"It addresses noisy and irrelevant intercompany relationships that can arise in prior graph attention approaches, as well as inefficiencies caused by computing attention coefficients separately across layers.\"},{\"question\":\"How does SGAT-SP reduce noisy edges in the stock graph?\",\"answer\":\"It assigns a binary mask to edges to decide whether an edge is used for node aggregation, suppressing noisy connections during training and inference.\"},{\"question\":\"What performance improvements does SGAT-SP report compared with ML-GAT?\",\"answer\":\"SGAT-SP achieves an average accuracy of 0.83 versus ML-GAT’s 0.827, and it reduces inference time by 85%, lowering computational expenses.\"}]","SGAT-SP: Sparse Graph Attention Network for Stock Prediction - Research Article | PDF"]