[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119227-en":3,"doc-seo-119227-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},119227,687197207639,"Asher","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","Financial market predictability with artificial intelligence and machine learning techniques - PhD thesis","This thesis investigates the intersection of financial markets and machine learning, aiming to improve financial return predictability and address missing values in financial datasets. Interpretable machine learning models are used to enhance understanding of market dynamics. LassoNet forecasts U.S. industry portfolio returns while enforcing covariate sparsity, improving forecasting accuracy and enabling profitable industry ETF portfolios with strong risk-adjusted performance. BRITS imputes missing hedge fund returns, boosting forecast accuracy and investment returns. TabNet predicts excess-return direction, surpassing alternatives and capturing valuation, lagged effects, and seasonal and cross-industry relationships.","Financial market predictability with artificial intelligence and machine learning techniques  \nLazaros Zografopoulos  \nA thesis submitted for the degree of PhD  \nat the University of St Andrews  \n2025  \nFull metadata for this item is available in St Andrews Research Repository at:  \n[https://research-repository.st-andrews.ac.uk/](https://research-repository.st-andrews.ac.uk/)  \n[Identifier to use to cite or link to this thesis:](Identifier to use to cite or link to this thesis:)[ ](Identifier to use to cite or link to this thesis:)DOI: [https://doi.org/10.17630/sta/1211](https://doi.org/10.17630/sta/1211)  \nThis item is protected by original copyright  \nABSTRACT  \nThis thesis explores the intersection of financial markets and machine learning, focusing on financial return predictability and the imputation of missing values in financial datasets. The research aims to enhance our understanding of financial market dynamics through the lens of interpretable machine learning models. Specifically, the thesis employs advanced machine learning techniques to predict financial returns and address missing data issues, which are common but often overlooked in financial literature.  \nThe first chapter uses an interpretable machine learning model, LassoNet, to forecast U.S. industry portfolio returns. LassoNet combines a regularization mechanism with a neural network architecture to enforce covariate sparsity. The findings show that LassoNet outperforms linear and non-linear models in forecasting accuracy, with valuation ratios and individual and cross-industry lagged returns being the most critical covariates. The model's forecasts enable the construction of profitable industry ETF portfolios that outperform benchmarks in annualized returns, Sharpe ratios, and alpha values.  \nThe second chapter focuses on imputing missing hedge fund return data using a deep learning model, the bidirectional recurrent imputation network for time series (BRITS) . BRITS is compared to other imputation methods like the cross-sectional mean and matrix completion. The results indicate that BRITS significantly enhances forecasting accuracy and economic performance of predictive models when used to impute the missing values in the data. The imputed data leads to lower out-ofsample errors and higher investment returns, demonstrating BRITS' superiority in handling missing values.  \nIn the third chapter, the state-of-the-art neural network architecture TabNet is utilized to forecast the directional movements of excess returns in industry portfolios. TabNet surpasses other models in classification accuracy and highlights the importance of valuation ratios and lagged returns in its predictions. The model effectively captures seasonal effects and cross-industry economic links and attains the highest annualized returns and positive Sharpe ratios in trading applications.  \nI dedicate this thesis to my mother, Anta Ptochopoulou, my brother, Dr. Ioannis Zografopoulos,  \nand to my late father, Fotis Zografopoulos, who never saw this adventure  \nACKNOWLEDGEMENTS  \nFirst and foremost, I would like to express my deepest gratitude to my supervisor Dr. Ioannis Psaradellis. Your guidance and unwavering support have been invaluable. Your expertise and constructive comments have greatly enhanced the quality of my work. I am also thankful to Dr. Maria Chiara Iannino for her insightful feedback and support throughout my Ph. D. studies. I would like to thank the School of Economics and Finance for fostering a supportive and dynamic research environment.  \nI am also grateful to Sofoklis Achillopoulos Foundation for providing me with a scholarship for the third and fourth years of the Ph. D. programme. This scholarship has been crucial in completing my studies and producing high-quality research. I would also like to thank the Clelia Hajiioannou Foundation for generously offering me a scholarship during my second year of studies.  \nMany thanks to my colleagues and fellow students in the ","cbCaikZe08cLgW2A","https://ap.wps.com/l/cbCaikZe08cLgW2A","pdf",3609092,1,186,"English","en",105,"# Abstract\n# Chapter 1: LassoNet for U.S. industry portfolio returns\n## Feature selection and covariate sparsity\n## Forecasting performance and portfolio construction\n# Chapter 2: BRITS for missing hedge fund return imputation\n## Comparison with baseline imputation methods\n## Impact on forecasting errors and investment returns\n# Chapter 3: TabNet for directional excess-return forecasts\n## Classification accuracy and key predictors\n## Trading implications: seasonal and cross-industry links","[{\"question\":\"What problems does the thesis focus on in financial markets?\",\"answer\":\"It focuses on predicting financial returns and imputing missing values in financial datasets, using interpretable machine learning approaches to better capture market dynamics.\"},{\"question\":\"How does LassoNet improve return forecasting in the first chapter?\",\"answer\":\"LassoNet combines regularization with a neural network to enforce covariate sparsity, producing more accurate forecasts than linear and non-linear alternatives and identifying key predictors such as valuation ratios and lagged returns.\"},{\"question\":\"How is missing hedge fund data handled, and what is the benefit?\",\"answer\":\"The thesis uses BRITS, a deep learning time-series imputation model, compared against methods like cross-sectional means and matrix completion. BRITS reduces out-of-sample errors and improves investment returns when used in predictive models.\"}]","Financial market predictability with artificial intelligence and machine learning techniques - PhD thesis | PDF",1785723191,469,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"financial-market-predictability-with-artificial-intelligence-and-machine-learning-techniques-phd-thesis","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/financial-market-predictability-with-artificial-intelligence-and-machine-learning-techniques-phd-thesis/119227/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problems does the thesis focus on in financial markets?","Question",{"text":75,"@type":76},"It focuses on predicting financial returns and imputing missing values in financial datasets, using interpretable machine learning approaches to better capture market dynamics.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does LassoNet improve return forecasting in the first chapter?",{"text":80,"@type":76},"LassoNet combines regularization with a neural network to enforce covariate sparsity, producing more accurate forecasts than linear and non-linear alternatives and identifying key predictors such as valuation ratios and lagged returns.",{"name":82,"@type":73,"acceptedAnswer":83},"How is missing hedge fund data handled, and what is the benefit?",{"text":84,"@type":76},"The thesis uses BRITS, a deep learning time-series imputation model, compared against methods like cross-sectional means and matrix completion. BRITS reduces out-of-sample errors and improves investment returns when used in predictive models.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]