[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120617-en":3,"doc-seo-120617-105":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":20,"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},120617,13056703020460,"Valentina","https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923",8,"Research & Report","A Machine-Learning Model for Predicting the SOFR Term-structure - A Thesis","The thesis develops a comprehensive machine-learning framework for modeling and forecasting the SOFR term structure, motivated by SOFR’s role as the leading benchmark for U.S.-dollar interest rate derivatives. It constructs a daily piecewise-constant SOFR forward rate curve using the official CME methodology and introduces policy-driven discontinuities at FOMC dates. The curve is fitted to the dynamic Nelson–Siegel model to extract level, slope, and curvature time series, then modeled with RNNs including LSTM, with Kalman-filter state-space estimation. The approach is evaluated via out-of-sample forecasts across loss functions and benchmarked against curve persistence, showing robust short-term forecasting while identifying challenges around abrupt monetary policy changes.","A Machine Learning Model for Predicting the SOFR Term-structure  \nYiming Guo  \nA Thesis  \nin the Department  \nof  \nMathematics and Statistics  \nPresented in Partial Fulfillment of the Requirements  \nfor the Degree of  \nMaster of Science in Mathematics and Statistics at Concordia University  \nMontr´eal, Qu´ebec, Canada  \nAugust 2025  \n© Yiming Guo, 2025  \nCONCORDIA UNIVERSITY School of Graduate Studies  \nThis is to certify that the thesis prepared  \nBy: Mr. Yiming Guo  \nEntitled: A Machine-Learning Model for Predicting SOFR Term-Structure  \nand submitted in partial fulfillment of the requirements for the degree of  \nMaster of Science  \ncomplies with the regulations of the University and meets the accepted standards with respect to originality and quality.  \nSigned by the final examining committee:  \n  Chair  \nDr. Frédéric Godin  \n  Examiner  \nDr. Patrice Gaillardetz  \n  Examiner  \n  7KHVLV Supervisor􀀋V􀀌 Dr. Cody Hyndman  \nBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBBB 7KHVLV 6XSHUYLVRU􀀋V􀀌  \nApproved by  BBBB  Dr. Lea Popovic Chair of Department or Graduate Program Director  \nDr. Pascale Sicotte 'HDQ RI Faculty of Arts & Science  \nABSTRACT  \nA Machine Learning Model for Predicting the SOFR Term-structure  \nYiming Guo, M.Sc.  \nConcordia University, 2025  \nThe Secured Overnight Financing Rate (SOFR) has emerged as the leading benchmark for U.S. dollar-denominated interest rate derivatives, replacing LIBOR due to its transparency and robustness. This thesis develops a comprehensive framework for modeling and forecasting the SOFR term structure using machine learning methods, with a particular focus on CME SOFR futures. We first apply the official CME methodology to construct a daily, piecewise-constant SOFR forward rate curve, incorporating policy-driven discontinuities at Federal Open Market Committee (FOMC) dates. The curve is then fitted to the dynamic Nelson–Siegel (DNS) model, extracting time series of level, slope, and curvature factors. To capture and predict the evolution of these factors, we implement recurrent neural networks (RNNs), including Long Short-Term Memory (LSTM) architectures, and integrate a Kalman filter for state-space estimation. The performance of the proposed model is evaluated through out-of-sample forecasts under various loss functions, including mean squared error (MSE) and mean absolute error (MAE), and benchmarks against curve persistence. Our results show that the machine learning approach provides robust short-term forecasts for the SOFR term structure. However, accurately modeling and forecasting abrupt changes around monetary policy announcements remains a challenge, highlighting an important direction for future research. This research offers a practical and robust modeling strategy for interest rate risk management, with direct applications in the pricing and risk assessment of SOFR-linked financial products.  \nKeywords: SOFR, term structure modeling, CME SOFR Futures, dynamic Nelson-Siegel, Kalman filter, recurrent neural network, long short-term memory, FOMC, yield curve forecasting, machine learning, interest rate derivatives.  \nAcknowledgments  \nFirst, I would like to express my deepest gratitude to my supervisor, Dr. Cody Hyndman, for his clear guidance, encouragement, and invaluable advice throughout the coursework and research. I also thank my parents for their support and encouragement of my decision to pursue a second undergraduate degree in mathematics and then a master’s degree. I am grateful to my grandparents for their deep love. Finally, I am thankful to my best friends Crystal and Jeffrey for their constant support and companionship.  \nContents  \nList of Figures v  \nList of Tables vi  \nList of Variables viii  \n1 SOFR Futures 6  \n1.1 One-Month SOFR Futures (SR1) ........................ 6  \n1.2 Three-Month SOFR Futures (SR3) ....................... 9  \n2 CME Term SOFR Reference Rates Methodology 13  \n2.1 Input Data .................................... 13  \n2.2 Calculation Methodology ...","cbCaiueT9Y3vFMKV","https://ap.wps.com/l/cbCaiueT9Y3vFMKV","pdf",3348163,1,68,"English","en",105,"# Abstract\n# SOFR Futures\n## One-Month SOFR Futures (SR1)\n## Three-Month SOFR Futures (SR3)\n# CME Term SOFR Reference Rates Methodology\n## Input Data\n## Calculation Methodology\n## Broyden-Fletcher-Goldfarb-Shanno (BFGS) Optimization\n## Replication of the CME Methodology\n# Predictive Machine Learning Model with the Kalman Filter\n## Dynamic Nelson-Siegel Term Structure\n## Recurrent Neural Network Modeling\n### Long Short-Term Memory (LSTM)\n### Input layer\n### State layer\n### Residual layer\n### Kalman filter layer\n### Loss function\n# Numerical Implementation\n# Conclusion and Future Research","[{\"question\":\"How does the thesis construct the SOFR forward rate curve?\",\"answer\":\"It applies the official CME methodology to build a daily, piecewise-constant SOFR forward rate curve, incorporating policy-driven discontinuities at FOMC dates.\"},{\"question\":\"What models are used to represent the SOFR term structure dynamics?\",\"answer\":\"The curve is fitted to the dynamic Nelson–Siegel (DNS) model to obtain time series for level, slope, and curvature factors, which are then modeled using recurrent neural networks such as LSTM.\"},{\"question\":\"How is the predictive model evaluated and what are the main findings?\",\"answer\":\"Forecast performance is tested out of sample using loss functions like MSE and MAE and benchmarked against curve persistence. Results indicate robust short-term forecasts, while accurately capturing abrupt changes around monetary policy announcements remains challenging.\"}]","A Machine-Learning Model for Predicting the SOFR Term-structure - A Thesis | PDF",1785730924,171,{"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},"a-machine-learning-model-for-predicting-the-sofr-term-structure-a-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/a-machine-learning-model-for-predicting-the-sofr-term-structure-a-thesis/120617/",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":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"How does the thesis construct the SOFR forward rate curve?","Question",{"text":75,"@type":76},"It applies the official CME methodology to build a daily, piecewise-constant SOFR forward rate curve, incorporating policy-driven discontinuities at FOMC dates.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What models are used to represent the SOFR term structure dynamics?",{"text":80,"@type":76},"The curve is fitted to the dynamic Nelson–Siegel (DNS) model to obtain time series for level, slope, and curvature factors, which are then modeled using recurrent neural networks such as LSTM.",{"name":82,"@type":73,"acceptedAnswer":83},"How is the predictive model evaluated and what are the main findings?",{"text":84,"@type":76},"Forecast performance is tested out of sample using loss functions like MSE and MAE and benchmarked against curve persistence. Results indicate robust short-term forecasts, while accurately capturing abrupt changes around monetary policy announcements remains challenging.","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"]