[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127676-en":3,"doc-seo-127676-105":31,"detail-sidebar-cat-0-en-105":96},{"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":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},127676,962084925636,"Olivia Brown","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","Advances in Machine Learning Algorithms for Financial Risk Management - Doctoral Thesis","This doctoral thesis presents three novel machine learning techniques targeting interconnected financial risk management problems: credit risk classification, financial asset volatility forecasting, and portfolio optimization. A Hybrid Dual-Resampling and Cost-Sensitive approach mitigates class imbalance by generating synthetic minority samples with Gaussian mixture modelling, clustering the majority via k-means, selecting features with Extra Tree Ensemble, and using cost-sensitive logistic regression for default probability prediction. A Triple Discriminator GAN combined with a continuous wavelet transform decomposes volatility into signal-like and noise-like components. Finally, model-free deep reinforcement learning with a deep capsules network uses a Markov Differential Sharpe Ratio reward and a Multi-Memory Weight Reservoir to sequentially rebalance portfolios.","SCHOOL OF MATHEMATICAL, PHYSICAL AND COMPUTATIONAL SCIENCES  \nAdvances in Machine Learning Algorithms for Financial Risk Management  \nby  \nEmmanuel Osei-Brefo  \nThesis submitted for the degree of Doctor of Philosophy  \nPhD Computer Science  \nDepartment of Computer Science  \nNovember 2023  \nUNIVERSITY OF READING  \nDeclaration of original authorship  \nDeclaration: I confirm that this is my own work and the use of all material from other sources have been properly and fully acknowledged.  \nEmmanuel Osei-Brefo  \nAcknowledgements  \nI would like to express my deepest gratitude and appreciation to my main supervisor, Profes  \nsor Xia Hong for her enthusiasm, guidance, unwavering support, and invalua [U+FFFD][U+FFFD][U+FFFD +-.*+  \n0   0 . 1b0le expertise throughout my PhD journey. Her insights, constructive feedback, and encouragement have been instrumental in shaping the direction of this work. I would also like to thank my second supervisor, Professor Richard Mitchel for his inputs and proof-reads. I express my deepest gratitude and appreciation to my family members especially my wife Sabina, my mother, Mary Boahin and children; Declan, Perez, Mary-Anne and Emmanuel Jnr whose love and understanding have been the foundation upon which my academic pursuits have been built. Their belief in my abilities and the sacrifices they made have been a constant source of motivation, and I am forever grateful for their support. I also extend my gratitude to my fellow researchers and colleagues for their engaging discussions and collaborative efforts that made the research environment stimulating and inspiring. I thank the faculty members, administrative staff and support personnel at the Department of Computer Science at the University of Reading for fostering a conducive academic environment and providing the resources necessary for the successful completion of this thesis. I also thank my friends who have stood by me during the highs and lows of this journey for their unwavering friendship especially, Thomas Asante Mensah, Baaba Amoah and Dr. Kwaku Aduse-Poku for their help and advice. I acknowledge the collective effort of all those who have been involved in this endeavour, directly or indirectly.  \nAbstract  \nIn this thesis, three novel machine learning techniques are introduced to address distinct yet interrelated challenges involved in financial risk management tasks. These approaches collectively offer a comprehensive strategy, beginning with the precise classification of credit risks, advancing through the nuanced forecasting of financial asset volatility, and ending with the strategic optimisation of financial asset portfolios.  \nFirstly, a Hybrid Dual-Resampling and Cost-Sensitive technique has been proposed to combat the prevalent issue of class imbalance in financial datasets, particularly in credit risk assessment. The key process involves the creation of heuristically balanced datasets to effectively address the problem. It uses a resampling technique based on Gaussian mixture modelling to generate a synthetic minority class from the minority class data and concurrently uses k-means clustering on the majority class. Feature selection is then performed using the Extra Tree Ensemble technique. Subsequently, a cost-sensitive logistic regression model is then applied to predict the probability of default using the heuristically balanced datasets. The results underscore the effectiveness of our proposed technique, with superior performance observed in comparison to other imbalanced preprocessing approaches. This advancement in credit risk classification lays a solid foundation for understanding individual financial behaviours, a crucial first step in the broader context of financial risk management.  \nBuilding on this foundation, the thesis then explores the forecasting of financial asset volatility, a critical aspect of understanding market dynamics. A novel model that combines a Triple Discriminator Generative Adversarial Network with","cbCaidTJfNiOlPxP","https://ap.wps.com/l/cbCaidTJfNiOlPxP","pdf",7712924,2,1,249,"English","en",105,"# Abstract\n# Credit risk classification\n## Hybrid Dual-Resampling and Cost-Sensitive learning\n# Volatility forecasting\n## Triple Discriminator GAN with continuous wavelet transform\n# Portfolio optimization\n## Model-free reinforcement learning with deep capsules and Sharpe-ratio reward","[{\"question\":\"What are the three main contributions of the thesis in financial risk management?\",\"answer\":\"The thesis introduces techniques for (1) credit risk classification, (2) volatility time-series forecasting, and (3) financial asset portfolio optimization using reinforcement learning.\"},{\"question\":\"How does the thesis address class imbalance in credit risk datasets?\",\"answer\":\"It proposes a Hybrid Dual-Resampling and Cost-Sensitive technique that creates heuristically balanced datasets using Gaussian mixture modelling for synthetic minority samples and k-means clustering for majority instances, followed by feature selection and cost-sensitive logistic regression.\"},{\"question\":\"How is volatility decomposed and forecasted using the proposed model?\",\"answer\":\"A model combining a Triple Discriminator Generative Adversarial Network with a continuous wavelet transform decomposes volatility into signal-like and noise-like frequency components, enabling monitoring of non-stationary volatility.\"},{\"question\":\"What method is used for portfolio optimization and what reward guides learning?\",\"answer\":\"Portfolio optimization uses model-free deep reinforcement learning with a deep capsules network, guided by a Markov Differential Sharpe Ratio reward function and supported by a Multi-Memory Weight Reservoir for sequential rebalancing.\"}]","Advances in Machine Learning Algorithms for Financial Risk Management - Doctoral Thesis | PDF",1785940710,627,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":91,"head_meta":93,"extra_data":95,"updated_unix":29},"advances-in-machine-learning-algorithms-for-financial-risk-management-doctoral-thesis","",{"@graph":37,"@context":90},[38,54,69],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,48,51],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":20},"https://docshare.wps.com/document/","Document",{"item":49,"name":12,"@type":44,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":44,"position":53},"https://docshare.wps.com/document/advances-in-machine-learning-algorithms-for-financial-risk-management-doctoral-thesis/127676/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-22","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82,86],{"name":73,"@type":74,"acceptedAnswer":75},"What are the three main contributions of the thesis in financial risk management?","Question",{"text":76,"@type":77},"The thesis introduces techniques for (1) credit risk classification, (2) volatility time-series forecasting, and (3) financial asset portfolio optimization using reinforcement learning.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does the thesis address class imbalance in credit risk datasets?",{"text":81,"@type":77},"It proposes a Hybrid Dual-Resampling and Cost-Sensitive technique that creates heuristically balanced datasets using Gaussian mixture modelling for synthetic minority samples and k-means clustering for majority instances, followed by feature selection and cost-sensitive logistic regression.",{"name":83,"@type":74,"acceptedAnswer":84},"How is volatility decomposed and forecasted using the proposed model?",{"text":85,"@type":77},"A model combining a Triple Discriminator Generative Adversarial Network with a continuous wavelet transform decomposes volatility into signal-like and noise-like frequency components, enabling monitoring of non-stationary volatility.",{"name":87,"@type":74,"acceptedAnswer":88},"What method is used for portfolio optimization and what reward guides learning?",{"text":89,"@type":77},"Portfolio optimization uses model-free deep reinforcement learning with a deep capsules network, guided by a Markov Differential Sharpe Ratio reward function and supported by a Multi-Memory Weight Reservoir for sequential rebalancing.","https://schema.org",{"og:url":52,"og:type":92,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":94,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":97},[98,102,106,110,115,120,125,128,133,136,140],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":99,"show_sort_weight":100,"slug":101},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":103,"show_sort_weight":104,"slug":105},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":47,"category_name":107,"show_sort_weight":108,"slug":109},"Exam",70,"exam",{"id":111,"doc_module":4,"doc_module_name":47,"category_name":112,"show_sort_weight":113,"slug":114},5,"Comic",60,"comic",{"id":116,"doc_module":4,"doc_module_name":47,"category_name":117,"show_sort_weight":118,"slug":119},6,"Technology",50,"technology",{"id":121,"doc_module":4,"doc_module_name":47,"category_name":122,"show_sort_weight":123,"slug":124},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":126,"slug":127},30,"research-report",{"id":129,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":131,"slug":132},9,"Religion & Spirituality",20,"religion-spirituality",{"id":131,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":131,"slug":135},"World Cup","world-cup",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":137,"slug":139},10,"Lifestyle","lifestyle",{"id":141,"doc_module":4,"doc_module_name":47,"category_name":142,"show_sort_weight":111,"slug":143},19,"General","general"]