Original title: Non-Linear Classification as a Tool for Predicting Tennis Matches
Translated title: Non-Linear Classification as a Tool for Predicting Tennis Matches
Authors: Hostačný, Jakub ; Baniar, Matúš (advisor) ; Krištoufek, Ladislav (referee)
Document type: Master’s theses
Year: 2018
Language: eng
Abstract: [eng] [cze]
Charles University Faculty of Social Sciences Institute of Economic Studies MASTER'S THESIS Non-Linear Classification as a Tool for Predicting Tennis Matches Author: Be. Jakub Hostacny Supervisor: RNDr. Matus Baniar Academic Year: 2017/2018 Abstract In this thesis, we examine the prediction accuracy and the betting performance of four machine learning algorithms applied to men tennis matches - penalized logistic regression, random forest, boosted trees, and artificial neural networks. To do so, we employ 40 310 ATP matches played during 1/2001-10/2016 and 342 input features. As for the prediction accuracy, our models outperform current state-of-art models for both non-grand-slam (69%) and grand slam matches (79%). Concerning the overall accuracy rate, all model specifications beat backing a better-ranked player, while the majority also surpasses backing a bookmaker's favourite. As far as the betting performance is concerned, we develop six profitable betting strategies for betting on favourites applied to non-grand-slam with ROI ranging from 0.8% to 6.5%. Also, we identify ten profitable betting strategies for betting on favourites applied to grand slam matches with ROI fluctuating between 0.7% and 9.3%. We beat both bench­ mark rules - backing a better-ranked player as well as backing a bookmaker's... Charles University Faculty of Social Sciences Institute of Economic Studies MASTER'S THESIS Non-Linear Classification as a Tool for Predicting Tennis Matches Author: Be. Jakub Hostacny Supervisor: RNDr. Matus Baniar Academic Year: 2017/2018 Abstrakt V tejto diplomovej práci skúmame predikčnú přesnost' a výkon pri stávkovaní u štyroch strojovo učiacich sa algoritmov - penalizovaná logistická regresia, náhodný les, posilněné stromy a neuronové siete. Pri práci využíváme 40 310 ATP zápasov hraných počas obdobia 1/2001-10/2016. Co sa týká predikčnej přesnosti, naše modely prekonávajú najlepšie modely súčasnosti pre prediko- vanie negrandslamových (69%) ako aj modely pre predikovanie grandslamových zápasov (79%). Všetky špecifikácie modelov sú presnejšie ako predikovanie na základe rebríčkového postavenia hráčov, zatial'čo váčšina specifikách je přes­ nějších ako predikovanie na základe vypísaných kurzov stávkových kancelárií. Co sa týká návratnosti pri stávkovaní, vytvořili sme šest'profitabilných stratégií pre stávkovanie na favoritov pre negrandslamové zápasy (návranosť investície v rozmedzí 0.8-6.5%). Taktiež sme identifkovali desať profitabilných stratégií pre stávkovanie na favoritov pre grandslamové zápasy (návranosť investície v rozmedzí 0.7-9.3%). Naše modely prinášajú vyššiu návratnost' ako stávkovanie...
Keywords: Feature construction; Neural networks; SVM; Neuronové sítě; Podpůrné vektory

Institution: Charles University Faculties (theses) (web)
Document availability information: Available in the Charles University Digital Repository.
Original record: http://hdl.handle.net/20.500.11956/94871

Permalink: http://www.nusl.cz/ntk/nusl-373014


The record appears in these collections:
Universities and colleges > Public universities > Charles University > Charles University Faculties (theses)
Academic theses (ETDs) > Master’s theses