[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120841-en":3,"doc-seo-120841-105":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},120841,1649267921044,"Ava Thompson","https://us-avatar.wpscdn.com/avatar/1800007509477c92dfb?_k=1782875107921204101",8,"Research & Report","MACHINE LEARNING AND ASSET MANAGEMENT - CLUSTERING FOR PORTFOLIO CONSTRUCTION - Portfolio construction research","This work project explores how machine learning supports portfolio management through an experimental asset allocation approach. Clustering is used to reduce similarities between clusters, aiming to improve diversification benefits for portfolio decision-makers. The allocation strategy applies K-Means clustering to stock prices using the 100 largest-capitalization stocks in the S&P 500 index, with full backtesting and performance measurement over the considered period. Results show meaningful out-of-sample explanatory power, while also highlighting the difficulty of delivering consistently high abnormal returns.","A Work Project, presented as part of the requirements for the Award of a Master’s degree in Finance from the Nova School of Business and Economics.  \nMACHINE LEARNING AND ASSET MANAGEMENT: CLUSTERING FOR  \nPORTFOLIO CONSTRUCTION  \nGIACOMO VIALETTO  \nWork project carried out under the supervision of:  \nDaniele d’Arienzo  \nAbstract (100 words maximum)  \nThis research investigates how machine learning can be applied to portfolio management and the results ofa related experimental asset allocation. Clustering aims at minimizing inter-clustering similarities, therefore translating in potentially higher diversification benefits, one of the goals of portfolio managers. The chosen allocation strategy of this research is K-Means clustering on prices with the 100 stocks with largest capitalization in the S&P500 index, fully backtested and measured as of performance. The strategy yields interesting out-of-sample explanatory power, with good results over the considered period, although confirming the difficulty for portfolio managers to consistently deliver high abnormal returns.  \nKeywords (minimum of four)  \nMachine learning, asset management, clustering, portfolio  \nThis work used infrastructure and resources funded by Fundação para a Ciência e a Tecnologia (UID/ECO/00124/2013, UID/ECO/00124/2019 and Social Sciences DataLab, Project 22209), POR Lisboa (LISBOA-01-0145-FEDER-007722 and Social Sciences DataLab, Project 22209) and POR Norte (Social Sciences DataLab, Project 22209).  \n1. Machine Learning for financial applications: an overview  \n2. Concept of diversification  \n2.1. Benefits of diversification  \n2.2. Markowitz’s efficient frontier: main features and limitations  \n3. Unsupervised learning methods: clustering algorithm  \n3.1. Proximity matrix and concept of distance  \n3.2. Partitional clustering and hierarchical clustering  \n3.3. Main clustering models  \n3.3.1 Connectivity models  \n3.3.2 Centroids models  \n3.3.3 Distribution-based and density-based models  \n3.4. Number of clusters  \n4. Experimental results in Python: K-Means clustering on stock prices  \n4.1 Construction  \n4.2 Discussion  \n5. Conclusion  \n6. References  \n7. Appendix  \n1. Machine Learning for financial applications: an overview  \nAs the need for larger data sets emerge over decades, an efficient answer can be provided by machine learning techniques. The need for more efficient techniques of data management stems from several reasons: among them, that of responding to complex (high-dimensional) problems and that of handling an increasingly large amount of data, whose connections are increasingly hard to spot by means of classical statistical properties such as simple correlation (as demonstrated later) or methods of linear regression. Machine learning techniques can therefore provide an answer to these issues as they define methods that leverage data in order to improve performance on some datasets (Mitchell 1997) . As such, machine learning techniques identify as a highly interdisciplinary field, that links aspects of computer and data science, linear algebra, and statistical learning, among others. Applications to the financial sector are recent and mainly relate to sectors like portfolio management, price prediction, trade execution and credit rating. Hence, financial machine learning presents distinguishable features that are worth investigating, and next steps in research within this field can open opportunities to relevant applications. Oneof the fields where advances in financial machine learning are most tangible consists in portfolio management and construction. Most of the data sets available to portfolio managers are:  \n- High-dimensional (i.e., anywhere from a few dozens to many thousands of dimensions can be required to define a point or value within a dataset)  \n- Sparse (i.e., most of the elements can be equal to zero) (Di, Tao e Liu 2017)  \n- Hierarchical (i.e., the data can be organized into a tree-structure)  \nThe necessity to deal with these fe","cbCailkt7Ia30bOW","https://ap.wps.com/l/cbCailkt7Ia30bOW","pdf",1004890,1,35,"English","en",105,"# Machine Learning for financial applications: an overview\n## Concept of diversification\n### Benefits of diversification\n## Markowitz’s efficient frontier: main features and limitations\n# Unsupervised learning methods: clustering algorithm\n## Proximity matrix and concept of distance\n## Partitional clustering and hierarchical clustering\n## Main clustering models\n### Connectivity models\n### Centroids models\n### Distribution-based and density-based models\n## Number of clusters\n# Experimental results in Python: K-Means clustering on stock prices\n## Construction\n## Discussion\n# Conclusion\n# References\n# Appendix","[{\"question\":\"How does the research use clustering to support portfolio construction?\",\"answer\":\"Clustering minimizes similarities between clusters, which can translate into higher diversification benefits for portfolio managers.\"},{\"question\":\"What dataset and clustering method are used in the experimental allocation strategy?\",\"answer\":\"The strategy uses K-Means clustering on stock prices for the 100 largest-capitalization stocks in the S\\u0026P 500 index, with results fully backtested and measured as performance.\"},{\"question\":\"What do the experimental results indicate about abnormal returns?\",\"answer\":\"The strategy provides interesting out-of-sample explanatory power with good period results, yet it confirms the challenge for portfolio managers to consistently achieve high abnormal returns.\"}]","MACHINE LEARNING AND ASSET MANAGEMENT - CLUSTERING FOR PORTFOLIO CONSTRUCTION - Portfolio construction research | PDF",1785732292,88,{"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},"machine-learning-and-asset-management-clustering-for-portfolio-construction-portfolio-construction-research","",{"@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/machine-learning-and-asset-management-clustering-for-portfolio-construction-portfolio-construction-research/120841/",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},"How does the research use clustering to support portfolio construction?","Question",{"text":75,"@type":76},"Clustering minimizes similarities between clusters, which can translate into higher diversification benefits for portfolio managers.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What dataset and clustering method are used in the experimental allocation strategy?",{"text":80,"@type":76},"The strategy uses K-Means clustering on stock prices for the 100 largest-capitalization stocks in the S&P 500 index, with results fully backtested and measured as performance.",{"name":82,"@type":73,"acceptedAnswer":83},"What do the experimental results indicate about abnormal returns?",{"text":84,"@type":76},"The strategy provides interesting out-of-sample explanatory power with good period results, yet it confirms the challenge for portfolio managers to consistently achieve high abnormal returns.","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"]