[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117126-en":3,"doc-seo-117126-105":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},117126,549758252649,"Ivy","https://ap-avatar.wpscdn.com/avatar/8000253669c5317157?_k=1778319167496531819",8,"Research & Report","Essays on Real Estate Finance and Machine Learning - Doctor of Philosophy","This PhD thesis presents three essays linking real estate finance and machine learning. It evaluates whether machine learning can improve out-of-sample return prediction for U.S. REITs, finding stronger predictability and substantial economic gains versus traditional OLS forecasts, and showing REITs to be more predictable than stocks. It then analyzes why, identifying distinct fundamental drivers, robustness to predictor set size for REITs, and persistent explanatory ability compared with stocks over time. The work also studies missing data methods, showing that absorbing observations with missing values improves prediction and can reverse marginal relationships, risking incorrect conclusions from smaller complete datasets.","ESSAYS ON REAL ESTATE FINANCE AND MACHINE  \nLEARNING  \nLEOWKAH SHIN  \nKing’s College  \nDepartment of Land Economy University of Cambridge  \nThis thesis is submitted for the degree of Doctor of Philosophy  \nMay 2024  \nDeclaration  \nThis thesis is the result of my own work and includes nothing which is the outcome of work done in collaboration except as declared in the preface and specified in the text. It isnot substantially the same as any work that has already been submitted, or, is being concurrently submitted, for any degree, diploma or other qualification at the University of Cambridge or any other University or similar institution except as declared in the preface and specified in the text. It does not exceed the prescribed word limit for the relevant Degree Committee.  \nAbstract  \nThis thesis presents three essays on the intersection of real estate finance and machine learning. Chapter One introduces my thesis. Chapter Two investigates if machine learning methods can deliver significant out-of-sample improvement in the return prediction of U.S. REITs. I find that return predictability is improved and REIT investors experience significant economic gains when using machine learning forecasts as compared to traditional OLS forecasts. I also discover that REITs are more predictable than stocks. Chapter Three takes a deep dive to understand why REITs differ from stocks in their predictability when using machine learning. I find that the fundamental drivers of predictability for REITs are very different from stocks, that REITs are robust to the size of the predictor set whereas stocks are not, and that the ability offactors to explain the return of stocks has declined over time but it has held constant for REITs. Chapter Four looks at how machine learning can be useful in real estate research when dealing with missing data. I demonstrate impressive gains in outof-sample predictions once I absorb observations with missing values. More importantly, I find that researchers may draw the wrong conclusions if they solely rely on complete but smaller data sets, as marginal relationships between independent and dependent variables can switch signs once models are permitted to absorb the much larger set of observations that contain missing values. Chapter Five concludes and wraps up my thesis.  \nAcknowledgements  \nI still remember stepping into the Department of Land Economy for the first time in September 2018 and meeting Thies Lindenthal. He was the course director of our MPhil programme in Real Estate Finance. He made an impression, literally. Besides being the tallest professor I have ever encountered in my life, he served as an inspiration to all of us in the MPhil course. Today, I am grateful to him for inspiring me to take up my PhD and guiding me through the journey as my supervisor.  \nI am also grateful to Brent Ambrose, Colin Lizieri, Philip Kalikman, Rongjie Zhang and participants of the 2023 AREUEA International Conference at the University of Cambridge for their useful comments and suggestions for the papers presented in this thesis. The same gratitude goes to Kelvin Wong, Chongyu Wang, Daniel Ruf, Wayne Wan, Derek Ho and participants of the 2023 Real Estate Finance and Investment Symposium at the University of Hong Kong for their invaluable feedback.  \nI wish to thank my wife for her love and support throughout my PhD journey. Additionally, I suffer from dad guilt where I feel like I am not doing enough for my daughters because of the PhD programme. I have resolved to spend more time on their education and in their daily lives upon the completion of my PhD. This thesis is dedicated to the three ladies in my life: Lijun, Raelynn and Raeanne.  \nContents  \n1 Introduction 7  \n2 Enhancing REIT Return Forecasts via Machine Learning 13  \n2.1 Introduction ........................................ 14  \n2.2 Literature review ..................................... 16  \n2.3 Data and methodology .................................. 17  \n2","cbCaiuEz2EEQ47CT","https://ap.wps.com/l/cbCaiuEz2EEQ47CT","pdf",3596614,1,182,"English","en",105,"# Introduction\n# Enhancing REIT Return Forecasts via Machine Learning\n## Literature review\n## Data and methodology\n## Empirical analysis\n## Portfolio analysis\n# Decomposing Predictability of REIT Returns\n## Literature review\n## Data and methodology\n## Empirical analysis\n# Machine Learning and Missing Data in Real Estate\n## Literature review","[{\"question\":\"How does the thesis measure whether machine learning improves REIT return forecasts?\",\"answer\":\"It tests out-of-sample return predictability for U.S. REITs and compares machine learning forecasts with traditional OLS forecasts to assess improvements and economic gains.\"},{\"question\":\"What explains differences in predictability between REITs and stocks?\",\"answer\":\"The thesis finds that the fundamental drivers differ, that REIT predictability is robust to the size of the predictor set, and that the explanatory ability of factors for stocks declines over time while it stays more constant for REITs.\"},{\"question\":\"What risk arises when researchers use only complete but smaller datasets with missing values?\",\"answer\":\"The thesis shows that marginal relationships between variables can switch signs once models are allowed to absorb much larger datasets containing missing values, leading to potentially wrong conclusions.\"}]","Essays on Real Estate Finance and Machine Learning - Doctor of Philosophy | PDF",1785674010,459,{"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},"essays-on-real-estate-finance-and-machine-learning-doctor-of-philosophy","",{"@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/essays-on-real-estate-finance-and-machine-learning-doctor-of-philosophy/117126/",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","2026-08-02",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 thesis measure whether machine learning improves REIT return forecasts?","Question",{"text":75,"@type":76},"It tests out-of-sample return predictability for U.S. REITs and compares machine learning forecasts with traditional OLS forecasts to assess improvements and economic gains.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What explains differences in predictability between REITs and stocks?",{"text":80,"@type":76},"The thesis finds that the fundamental drivers differ, that REIT predictability is robust to the size of the predictor set, and that the explanatory ability of factors for stocks declines over time while it stays more constant for REITs.",{"name":82,"@type":73,"acceptedAnswer":83},"What risk arises when researchers use only complete but smaller datasets with missing values?",{"text":84,"@type":76},"The thesis shows that marginal relationships between variables can switch signs once models are allowed to absorb much larger datasets containing missing values, leading to potentially wrong conclusions.","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"]