[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-290243-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-290243-en":131},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","machine-learning-mutual-fund-flows-working-paper-no-26-03","Machine learning mutual fund flows - Working Paper - No. 26-03","","This working paper improves out-of-sample predictability of future mutual fund flows using state-of-the-art machine learning. Nonlinear ML models substantially outperform linear specifications in out-of-sample R-squared. Interpretable ML methods highlight past flows and Morningstar ratings as the strongest predictors of netflows, while other past performance measures matter less. The effects of Morningstar ratings and expenses increase over time, and interactions between past flows and ratings materially affect future flows.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/machine-learning-mutual-fund-flows-working-paper-no-26-03/290243/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/machine-learning-mutual-fund-flows-working-paper-no-26-03/290243.png","ImageObject",300,407,{"name":42,"@type":43},"Finn","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-30","2026-09-17",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",12,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"Which models improve out-of-sample predictability of fund flows in this paper?","Question",{"text":63,"@type":64},"State-of-the-art nonlinear machine learning models improve out-of-sample predictability and significantly outperform linear models in out-of-sample R-squared.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What variables are identified as most important for predicting net flows?",{"text":68,"@type":64},"Interpretable ML results show that past flows and the Morningstar rating are the most important predictors of netflows.",{"name":70,"@type":61,"acceptedAnswer":71},"How do the authors connect ML-based predictions to fund performance differences?",{"text":72,"@type":64},"The results show that ML-based fund flow predictions can be used ex-ante to differentiate between high and low-performing mutual funds, with the most improvable predictions corresponding to the worst performance.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},290243,1789641102,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,127],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":25,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":25,"category_name":129,"show_sort_weight":97,"slug":130},19,"General","general",{"code":4,"msg":82,"data":132},{"doc_id":79,"user_id":133,"nickname":42,"user_avatar":134,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":86,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":80,"read_time":144,"preview_page_count":4},34359740700684,"https://ap-avatar.wpscdn.com/avatar/1f400023980c374ae676?_k=1777273430885731487","Fausch, Jürg; Frigg, Moreno; Ruenzi, Stefan; Weigert, Florian  \nWorking Paper  \nMachine learning mutual fund flows  \nCFR Working Paper, No. 26-03  \nProvided in Cooperation with:  \nCentre for Financial Research (CFR), University of Cologne  \nSuggested Citation: Fausch, Jürg; Frigg, Moreno; Ruenzi, Stefan; Weigert, Florian (2026) : Machine learning mutual fund flows, CFR Working Paper, No. 26-03, University of Cologne, Centre for Financial Research (CFR), Cologne  \nThis Version is available at:  \n[https://hdl.handle.net/10419/337467](https://hdl.handle.net/10419/337467)  \nStandard-Nutzungsbedingungen:  \nDie Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.  \nSie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.  \nSofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte.  \nTerms of use:  \nDocuments in EconStor maybe saved and copied foryour personal and scholarly purposes.  \nYou are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public.  \nIf the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence.  \nCFR Working Paper NO . 26-03 Machine Learning Mutual Fund Flows  \nJ . Fausch • M . Frigg •  \nS . Ruenzi • F . Weigert  \nMachine Learning Mutual Fund Flows  \nJürg Fausch∗ Moreno Frigg† Stefan Ruenzi‡ Florian Weigert §  \nThis draft: May 03, 2025  \nFirst draft: April 30, 2024  \nAbstract  \nWe present improved out-of-sample predictability of future fund flows using state-of-the-art machine learning methods. Nonlinear machine learning models significantly outperform linear models in terms of out-of-sample R-squared. Using interpretable ML methods, we identify past flows and the Morningstar rating as the most important predictors for netflows, while other past performance variables are of minor importance. We find that the importance of Morningstar ratings and expenses has increased over time. In addition, the interaction effect of past flows with the Morningstar rating has a substantial impact on future flows. Furthermore, our results demonstrate that machine learning-based fund flow predictions can be used to ex-ante differentiate between high and low-performing mutual funds. Finally, funds whose flow predictions can be improved the most using ML reveal the worst performance, consistent with the idea that liquidity management is particularly challenging for these funds.  \nJEL Classification: C45, C52, C53, C55, G10, G11, G12, G17, G23  \nKeywords: Machine learning, fund flow prediction, big data, interpretable machine learning  \n∗ Jürg Fausch is from the IFZ, Lucerne University of Applied Sciences and Arts, Suurstoffi 1, 6343 Rotkreuz, Switzerland. E-mail: juerg.fausch@hslu.ch.  \n†Moreno Frigg is from the Institute of Financial Analysis, University of Neuchâtel, Rue Abram-Louis-Breguet 2, 2000 Neuchâtel, Switzerland, and the IFZ, Lucerne University of Applied Sciences and Arts, Suurstoffi 1, 6343 Rotkreuz, [Switzerland. E-mail:](Switzerland. E-mail: moreno.frigg@unine.ch)[ moreno.frigg@unine.ch](Switzerland. E-mail: moreno.frigg@unine.ch).  \n‡Stefan Ruenzi is from the Department of Finance, University of Mannheim, L9, 1-2, 68161 Mannheim, Germany. E-mail: [stefan.ruenzi@uni-mannheim.de](stefan.ruenzi@uni-mannheim.de).  \n§ Florian Weigert is from the Institute of Financial Analysis, University of Neuchâtel, Rue Abram-LouisBreguet 2, 2000 Neuchâtel, Switzerland. E-mail: [florian.weigert@unine.ch](florian.w","cbCaivLjGPKF9BlT","https://ap.wps.com/l/cbCaivLjGPKF9BlT","pdf",2661743,"English","# Abstract\n# Introduction\n## Liquidity and unpredictability of fund flows\n## Motivation for machine learning methods\n## Predictability and nonlinear relationships","[{\"question\":\"Which models improve out-of-sample predictability of fund flows in this paper?\",\"answer\":\"State-of-the-art nonlinear machine learning models improve out-of-sample predictability and significantly outperform linear models in out-of-sample R-squared.\"},{\"question\":\"What variables are identified as most important for predicting net flows?\",\"answer\":\"Interpretable ML results show that past flows and the Morningstar rating are the most important predictors of netflows.\"},{\"question\":\"How do the authors connect ML-based predictions to fund performance differences?\",\"answer\":\"The results show that ML-based fund flow predictions can be used ex-ante to differentiate between high and low-performing mutual funds, with the most improvable predictions corresponding to the worst performance.\"}]","Machine learning mutual fund flows - Working Paper - No. 26-03 | PDF",227]