[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119075-en":3,"doc-seo-119075-105":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},119075,4398048950312,"Violet","https://ap-avatar.wpscdn.com/avatar/400002538284de19e3c?_k=1778320343897328908",8,"Research & Report","Predicting Tax Treaty Formation Using Machine Learning - Implications for Parliamentary Practice","The presentation analyzes tax treaty formation as a complex policy decision and outlines how machine learning can estimate double tax treaty formation probabilities for policy use. It reviews key determinants such as withholding tax rates, FDI allocations, transfer pricing and auditing costs, governance and tax haven bargaining power, and international relations including power asymmetries and information sharing. Using supervised binary classification with bilateral gravity features, it evaluates algorithms on training and test periods and highlights random forest performance and variable importance to support negotiation strategy and risk assessment for FDI diversion.","Predicting Tax Treaty FormationUsing Machine Learning:Implications for ParliamentaryPractice  \n·Dmitry Erokhin,PhD  \n·20 June 2024  \nErokhin,D.,&Zagler,M.(2023).Explaining and Predicting Double TaxTreaty Formation with Machine LearningAlgorithms.WU International TaxationResearch Paper Series,(2023-03).  \n# Tax treaty formation\n\n## ·Main goals of tax treaties:\n\no Boost trade and investment by eliminating tax barriers;  \no Fight tax evasion and double non-taxation with anti-avoidance measures;o Exchange of information,a primary focus of new tax treaties.  \n·Complex decision-making process precedes tax treaty formation:  \no Historically driven by \"chess-games between superpowers\",key persons'decisions,and corporate lobbyism;o Policy diffusion and other factors may influence tax treaty formation.  \n·Double tax treaty probability can be predicted using machine learning approach.  \n## ·Highly relevant for policymakers:\n\no Identifies advantageous treaties to pursue given limited capacity of negotiators;  \no Helps assess risk of FDI diversion to neighboring jurisdictions;  \no Aids in understanding level playing field for multinational firms in foreign markets.  \n# Literature-tax treaty formation\n\n## Economic factors:\n\n·Personal tax rates,non-resident withholding tax on dividends and interest  \n·FDI stock,symmetric allocation of FDI  \n·Transfer price,auditing costs,production costs  \n·Revenue sharing differences  \n## Governance and policy factors:\n\n·Tax haven bargaining power  \n·Good governance  \n·Experience in entering into tax treaties  \n## International relations factors:\n\n·Spatial spillovers and dependence in tax treatyformation  \n·Export-product competitors  \n·Power asymmetries between signatories  \n·Developed countries compensating developingcountries'tax base losses through foreign aid  \n## Other factors:\n\n·Common language  \n·Information sharing,tax audit,and revenue sharingagreements  \nWithholding tax rates:  \n·Positive relationship with withholding tax ratesnegotiated in past tax treaties  \n·Spatial dependence of dividends withholding tax rates  \n·Increase in withholding tax rates with asymmetric FDIactivities  \n# Machine learning in economics\n\n\"Machine learning\"economicssearches on Google scholar  \n140000  \n120000  \n100000  \n80000  \n·Uncovering generalizable patterns and findingfunctions with high out-of-sample predictivepower,which is important for policy makers.  \n60000  \n40000  \n·Limited prior theoretical assumptions or majorassumptions on variable distribution.  \n·Increased application in recent years.  \n·Various areas of economics such as energyeconomics,growth economics,crypto economics,urban economics,and taxation.  \n·Modeling complex and flexible relationships.  \n20000  \n0  \n20102015201920202021202220232024  \n# Methodology and data\n\n·Binary classification problem:country pairs as \"having tax treaties\"or \"not having taxtreaties\".  \n·c_ml_stata_cv command,Stata/Python integration,scikit-learn API.  \n·Supervised machine learning algorithms:classification tree,random forest,boosting,regularized multinomial,nearest neighbor,neural network,naive Bayes,support vectormachine,and standard multinomial algorithms.  \n·Classification error rate on the test data,sensitivity,precision,specificity,F1-score,and area-  \nunder-the-curve.  \n·Two periods:2018 for training the machine and 2019 for testing it.  \n·2800 country pairs with tax treaties and 6200 without tax treaties.  \n·29 gravity features constructed as bilateral variables.  \n# Algorithm selection\n\n·The selection of the most accurate algorithm isbased on the testing classification error rate,and the random forest algorithm has thelowest rate(0.057).  \n·The random forest algorithm outperforms otheralgorithms according to all the evaluationmetrics.  \n·Both the random forest algorithm and theclassification tree algorithm exhibit a zerotraining classification error rate and the lowesttwo testing classification error rates.  \n·A specific classification tree with 11 nodesdrawn using","cbCaioCvme0kolFd","https://ap.wps.com/l/cbCaioCvme0kolFd","pdf",798733,1,14,"English","en",105,"# Tax treaty formation\n## Main goals of tax treaties\n## Literature-tax treaty formation\n## Machine learning in economics\n# Methodology and data\n## Binary classification setup\n## Supervised algorithms and evaluation\n# Algorithm selection\n## Random forest performance and variable importance","[{\"question\":\"What main goals do tax treaties serve in the formation process?\",\"answer\":\"Tax treaties aim to boost trade and investment by reducing tax barriers and to combat tax evasion and double non-taxation through anti-avoidance measures. New treaties emphasize information exchange as a primary focus.\"},{\"question\":\"Which factors influence the formation of double tax treaties according to the literature review?\",\"answer\":\"Economic factors include personal tax rates, withholding taxes, FDI stock and allocation, transfer prices, auditing and production costs. Governance, experience, tax haven bargaining power, and international relations factors such as power asymmetries and information sharing also play a role.\"},{\"question\":\"How is machine learning applied to predict tax treaty formation and which algorithm performs best?\",\"answer\":\"The approach treats country pairs as a binary classification problem: having a tax treaty or not. Supervised algorithms are trained and tested across 2018 and 2019, evaluated with metrics like accuracy, sensitivity, precision, F1-score, and AUC, and random forest achieves the lowest testing classification error rate (0.057).\"}]","Predicting Tax Treaty Formation Using Machine Learning - Implications for Parliamentary Practice | PDF",1785722199,35,{"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},"predicting-tax-treaty-formation-using-machine-learning-implications-for-parliamentary-practice","",{"@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/predicting-tax-treaty-formation-using-machine-learning-implications-for-parliamentary-practice/119075/",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},"What main goals do tax treaties serve in the formation process?","Question",{"text":75,"@type":76},"Tax treaties aim to boost trade and investment by reducing tax barriers and to combat tax evasion and double non-taxation through anti-avoidance measures. New treaties emphasize information exchange as a primary focus.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which factors influence the formation of double tax treaties according to the literature review?",{"text":80,"@type":76},"Economic factors include personal tax rates, withholding taxes, FDI stock and allocation, transfer prices, auditing and production costs. Governance, experience, tax haven bargaining power, and international relations factors such as power asymmetries and information sharing also play a role.",{"name":82,"@type":73,"acceptedAnswer":83},"How is machine learning applied to predict tax treaty formation and which algorithm performs best?",{"text":84,"@type":76},"The approach treats country pairs as a binary classification problem: having a tax treaty or not. Supervised algorithms are trained and tested across 2018 and 2019, evaluated with metrics like accuracy, sensitivity, precision, F1-score, and AUC, and random forest achieves the lowest testing classification error rate (0.057).","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"]