[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-153060-105":59,"doc-detail-153060-en":131},{"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":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","stock-movement-prediction-a-multi-input-lstm-approach","Stock movement prediction - A multi-input LSTM approach","","Nonlinear and non-stationary financial time series create significant obstacles in stock movement prediction, often degrading traditional models’ efficiency and accuracy. To address this, the study proposes a hybrid framework combining wavelet transform (WT) with multi-input LSTM for predicting the trend of the SSE composite index. The method uses two stages: level-1 db4 decomposition to remove noise, then multi-source inputs including Chinese and US stock data and technical indicators. Experimental results show 72.19% accuracy and outperform single-input LSTM and several baseline machine learning models.",{"@graph":69,"@context":123},[70,84,106],{"@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/stock-movement-prediction-a-multi-input-lstm-approach/153060/",{"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/stock-movement-prediction-a-multi-input-lstm-approach/153060.png","ImageObject",300,407,{"name":92,"@type":93},"Evangeline","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-27","2026-08-27",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",13,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"Why is stock movement prediction difficult for existing models?","Question",{"text":113,"@type":114},"Financial time series are nonlinear and non-stationary with low signal-to-noise ratio, making traditional statistical assumptions less effective and lowering forecast accuracy.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How does the proposed WT and multi-input LSTM model work?",{"text":118,"@type":114},"It uses a two-stage design: level-1 wavelet (db4) decomposition to reduce noise, followed by a multi-input LSTM stage that leverages data from Chinese and US markets plus technical indicators.",{"name":120,"@type":111,"acceptedAnswer":121},"How does the proposed method perform compared with baseline approaches?",{"text":122,"@type":114},"The model achieves 72.19% accuracy, performing better than single-input LSTM and multiple baselines including decision tree, random forest, SVM, and XGBoost.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},153060,1787870490,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":105,"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":130,"read_time":144,"preview_page_count":4},13056703019662,"https://ap-avatar.wpscdn.com/avatar/be000253a8e92610077?_k=1778726343310543188","Received: 6 April 2021 Revised: 27 October 2022 Accepted: 4 January 2024  \nDOI: 10.1002/for.3071  \nRESE ARCH ARTICL E  \nStock movement prediction: A multi-input LSTM approach  \nPan Tang | Cheng Tang | Keren Wang  \nSchool of Economics and Management, Southeast University, Nanjing, 211189, China  \nCorrespondence  \nPan Tang, School of Economics and Management, Southeast University, Nanjing, Jiangsu 211189, China. [Email:](Email: pantang@seu.edu.cn)[ pantang@seu.edu.cn](Email: pantang@seu.edu.cn)  \nFunding information  \nNational Social Science Fund of China  \nAbstract  \nGenerally, the nonlinear and non-stationary financial time series becomes an obstacle in the process of stock movement prediction. But the recent theories of machine learning and deep learning have provided with some new solutions. Based on LSTM (long short-term memory), we propose a hybrid model of wavelet transform (WT) and multi-input LSTM to predict the trend of SSE composite index. It can mine valid data in time series and support different types of data as input. The whole model is divided into two stages. In the first stage, we adopt the level 1 decomposition with db4 mother wavelet to eliminate noise. In the second stage, combinative and qualitative analysis was made base on the data from Chinese stock market, US stock market, and technical indicators as input. According to the result, the proposed model, with the accuracy of 72.19%, performs better than single-input LSTM, decision tree, random forest, Support Vector Machine (SVM), and XGBoost.  \nKEYWOR DS  \ndeep learning, multi-input LSTM, stock movement prediction, wavelet transform  \n1 | INTRODUCTION  \n1.1 | Background  \nStock market is an important driving force for the economic development of a country. As China continuously accelerates the pace of capital market opening and deepens the overall reform, its stock market provides opportunities for both domestic and foreign investors to raise their wealth through investments. Nevertheless, stock market is pretty volatile and it could bring losses to investors. It is necessary to construct stock price prediction models to tackle market risks. Therefore, many scholars devote themselves to the research of stock price forecasting, which is also the research direction of this paper.  \nPredicting stock prices is a classic problem. Previous studies use traditional statistical methods in time series prediction, which can only predict stock prices, rather than price movement, such as GARCH (Agnolucci, 2009;  \nAwartani & Corradi, 2005; Bentes, 2015; Kriechbaumer et al., 2014), ARIMA (Torbat et al., 2018) . The methods are mainly based on linear, stable, and normal distribution assumptions. However, because the financial time series is a nonlinear and non-stationary time series with low SNR (signal-to-noise ratio), the traditional statistical methods are sometimes ineffective in prediction problem, which have poor efficiency and accuracy of the forecast results. Compared with statistical methods, machine learning (Patel et al., 2015a; Vijh et al., 2020 ; Nevasalmi, 2020) can handle noisy financial time series. And with the rapid development of artificial intelligence, machine learning stands out in this area. Nayak et al.(2015) combine Support Vector Machine (SVM) with K-Nearest Neighbor (KNN) to Indian stock market indices prediction. Henrique et al. (2018) use Support Vector Regression (SVR) on daily and up to the minute prices to predict stock prices in three different markets. They demonstrate that the mean square error caused by linear and radial kernels is smaller than that of random walk model,  \nJournal of Forecasting. 2024;43:1199–[1211.](1211. wileyonlinelibrary.com/journal/for)[ wileyonlinelibrary.com/journal/for](1211. wileyonlinelibrary.com/journal/for) © 2024 John Wiley & Sons Ltd.  1199  \nespecially when updating the model periodically. In particular, deep learning (Balaji et al., 2018; Chong et al., 2017; Liu & Long, 2020; Long et al., 2019) has robu","cbCait5S9f3kv4Ng","https://ap.wps.com/l/cbCait5S9f3kv4Ng","pdf",972947,"English","# Introduction\n## Background\n## Related work","[{\"question\":\"Why is stock movement prediction difficult for existing models?\",\"answer\":\"Financial time series are nonlinear and non-stationary with low signal-to-noise ratio, making traditional statistical assumptions less effective and lowering forecast accuracy.\"},{\"question\":\"How does the proposed WT and multi-input LSTM model work?\",\"answer\":\"It uses a two-stage design: level-1 wavelet (db4) decomposition to reduce noise, followed by a multi-input LSTM stage that leverages data from Chinese and US markets plus technical indicators.\"},{\"question\":\"How does the proposed method perform compared with baseline approaches?\",\"answer\":\"The model achieves 72.19% accuracy, performing better than single-input LSTM and multiple baselines including decision tree, random forest, SVM, and XGBoost.\"}]","Stock movement prediction - A multi-input LSTM approach | PDF",33]