[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117774-en":3,"doc-seo-117774-105":30,"detail-sidebar-cat-0-en-105":95},{"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},117774,1099513958762,"Logic","https://ap-avatar.wpscdn.com/avatar/1000023916a998db790?x-image-process=image/resize,m_fixed,w_180,h_180&k=1784791008015729253",8,"Research & Report","Behavioral Machine Learning - Computer Predictions of Corporate Earnings also Overreact - March 22, 2023","Machine-learning models can forecast corporate earnings better than humans in many financial contexts, yet the research rarely checks whether algorithmic predictions are more rational than human ones. This study evaluates earnings forecasts from linear regression and Gradient Boosted Regression Trees (GBRT) and finds GBRT’s average outperformance comes with systematic overreaction to news, violating rational expectations. Lowering the learning rate reduces overreaction but worsens out-of-sample accuracy. Analysts trained using machine-learning methods overreact less.","arXiv :2303 . 16158v1 [ q-fin . ST] 25 Mar 2023  \nBehavioral Machine Learning?  \nComputer Predictions of Corporate Earnings also Overreact *  \nMurray Z. Frank Jing Gao Keer Yang  \nMarch 22, 2023  \nAbstract  \nThere is considerable evidence that machine learning algorithms have better predictive abilities than humans in various 􀀛nancial settings. But, the literature has not tested whether these algorithmic predictions are more rational than human predictions. We study the predictions of corporate earnings from several algorithms, notably linear regressions and a popular algorithm called Gradient Boosted Regression Trees (GBRT). On average, GBRT outperformed both linear regressions and human stock analysts, but it still overreacted to news and did not satisfy rational expectation as normally de􀀛ned. By reducing the learning rate, the magnitude of overreaction can be minimized, but it comes with the cost of poorer out-of-sample prediction accuracy. Human stock analysts who have been trained in machine learning methods overreact less than traditionally trained analysts. Additionally, stock analyst predictions re􀀜ect information not otherwise available to machine algorithms.  \n* Respective a􀀞liations are Murray Frank, murra280@umn.edu, University of Minnesota, Minneapolis, MN. Jing  \nGao, [gao00268@umn.edu](gao00268@umn.edu), University of Minnesota, Minneapolis, MN. Keer Yang, [kkeyang@ucdavis.edu](kkeyang@ucdavis.edu), University of  \nCalifornia at Davis, Davis, CA  \nContents  \n1 Introduction 2  \n2 Data 4  \n3 Empirical Strategies 5  \n3.1 Using ML to Make Predictions ............................. 5  \n3.2 Testing for Overreaction ................................ 7  \n3.3 Predictable Forecast Errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8  \n4 Empirical Results 9  \n4.1 Prediction MSE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12  \n4.2 Machines Do Overreact . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14  \n4.3 Why Do Machine Overreact? .............................. 15  \n4.4 Overreaction to Aggregate or Firm-Speci􀀛c Shocks? ................. 18  \n4.5 Linear Regression Models Also Overreact ....................... 20  \n5 Stock Analysts 22  \n5.1 Technical Analysts ................................... 22  \n5.2 Technical VS Non-Tech Analysts ............................ 24  \n6 Model 28  \n6.1 Setup ........................................... 28  \n6.2 Equilibrium ....................................... 33  \n7 Conclusion 37  \nA Appendix 39  \n1 Introduction  \nAccording to Bordalo, Gennaioli and Shleifer (2022) there is strong evidence that the expectations of professional forecasters, corporate managers, consumers, and investors are o􀀘en biased towards overreaction. This evidence seriously challenges the rational expectations and e􀀞cient markets hypotheses. This bias is said to be rooted in human psychology, notably selective recall and a tendency to focus on salient information. This perspective can also help to explain certain 􀀛nancial and macroeconomic phenomena that are otherwise di􀀞cult to understand,( Bordalo et al., 2020, 2021a,b) .  \nWhile there is evidence that machine learning has better predictive power than humans in some 􀀛nance 􀀛nance settings, it is unclear whether machine algorithms produce more rational predictions than humans. To address this question, we study the predictions of corporate earnings from several algorithms, notably linear regressions1 and a popular algorithm called Gradient Boosted Regression Trees (GBRT), see Friedman (2001, 2002); Chen and Guestrin (2016) . We test for overreaction following the approach used in (Bordalo et al., 2021a) . On average, GBRT outperformed both linear regressions and human stock analysts. However, like human forecasts, the machine learning algorithm predictions also overreact to news and do not satisfy rational expectation as normally de􀀛ned.  \nIt is of interest to see if we can control the degree of overreac","cbCaiuhQS5eGedG5","https://ap.wps.com/l/cbCaiuhQS5eGedG5","pdf",810445,1,44,"English","en",105,"# Introduction\n## Data\n## Empirical Strategies\n### Using ML to Make Predictions\n### Testing for Overreaction\n### Predictable Forecast Errors\n# Empirical Results\n## Prediction MSE\n## Machines Do Overreact\n## Why Do Machine Overreact?\n## Overreaction to Aggregate or Firm-Specific Shocks?\n## Linear Regression Models Also Overreact\n# Stock Analysts\n## Technical Analysts\n## Technical VS Non-Tech Analysts\n# Model\n## Setup\n## Equilibrium\n# Conclusion\n# Appendix","[{\"question\":\"该研究比较了哪些预测方法来预测企业盈余？\",\"answer\":\"研究使用多种算法进行对比，重点包括线性回归以及梯度提升回归树（GBRT）。同时也将机器预测与人工股票分析师的预测进行比较。\"},{\"question\":\"GBRT 的预测结果体现了怎样的行为偏差？\",\"answer\":\"GBRT 在平均意义上优于线性回归和人类股票分析师，但仍会对新闻产生过度反应，且不满足通常定义下的理性预期。\"},{\"question\":\"如何缓解算法对新闻的过度反应？代价是什么？\",\"answer\":\"通过降低学习率可以最小化过度反应程度，但会降低算法对样本外数据的预测准确性。\"},{\"question\":\"机器学习训练的分析师与传统分析师相比有什么差异？\",\"answer\":\"具有机器学习背景训练的股票分析师对新闻的过度反应更小；研究还指出其预测包含了机器算法与标准数据库之外的额外信息。\"}]","Behavioral Machine Learning - Computer Predictions of Corporate Earnings also Overreact - March 22, 2023 | PDF",1785679485,111,{"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":90,"head_meta":92,"extra_data":94,"updated_unix":28},"behavioral-machine-learning-computer-predictions-of-corporate-earnings-also-overreact-march-22-2023","",{"@graph":36,"@context":89},[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/behavioral-machine-learning-computer-predictions-of-corporate-earnings-also-overreact-march-22-2023/117774/",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",[71,77,81,85],{"name":72,"@type":73,"acceptedAnswer":74},"该研究比较了哪些预测方法来预测企业盈余？","Question",{"text":75,"@type":76},"研究使用多种算法进行对比，重点包括线性回归以及梯度提升回归树（GBRT）。同时也将机器预测与人工股票分析师的预测进行比较。","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"GBRT 的预测结果体现了怎样的行为偏差？",{"text":80,"@type":76},"GBRT 在平均意义上优于线性回归和人类股票分析师，但仍会对新闻产生过度反应，且不满足通常定义下的理性预期。",{"name":82,"@type":73,"acceptedAnswer":83},"如何缓解算法对新闻的过度反应？代价是什么？",{"text":84,"@type":76},"通过降低学习率可以最小化过度反应程度，但会降低算法对样本外数据的预测准确性。",{"name":86,"@type":73,"acceptedAnswer":87},"机器学习训练的分析师与传统分析师相比有什么差异？",{"text":88,"@type":76},"具有机器学习背景训练的股票分析师对新闻的过度反应更小；研究还指出其预测包含了机器算法与标准数据库之外的额外信息。","https://schema.org",{"og:url":52,"og:type":91,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":93,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":96},[97,101,105,109,114,119,124,127,132,135,139],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":106,"show_sort_weight":107,"slug":108},"Exam",70,"exam",{"id":110,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},5,"Comic",60,"comic",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},6,"Technology",50,"technology",{"id":120,"doc_module":4,"doc_module_name":46,"category_name":121,"show_sort_weight":122,"slug":123},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":125,"slug":126},30,"research-report",{"id":128,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":130,"slug":131},9,"Religion & Spirituality",20,"religion-spirituality",{"id":130,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":130,"slug":134},"World Cup","world-cup",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":136,"slug":138},10,"Lifestyle","lifestyle",{"id":140,"doc_module":4,"doc_module_name":46,"category_name":141,"show_sort_weight":110,"slug":142},19,"General","general"]