Primjena strojnog učenja u predviđanju kretanja dionica i fondova

undergraduate thesis

• Title (english): Application of Machine Learning in Predicting Stock and Fund Movements
• Author: Dario Huić
• Mentor: Luka Humski (mentor)
• Committee member: Luka Humski (predsjednik povjerenstva)
• Committee member: Damir Pintar (član povjerenstva)
• Committee member: Ana Šarčević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-02, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovaj rad istražuje primjenu tehnika strojnog učenja, specifično stabla odluke, nasumične šume i strojeva s potpornim vektorima (SVM), u predviđanju kretanja cijena dionica i fondova. Korišten je skup podataka koji obuhvaća različite financijske instrumente, uključujući dionice tehnoloških tvrtki, energetskih kompanija te međunarodnih fondova. Podaci su prikupljeni iz pouzdanih izvora, transformirani i obogaćeni dodatnim značajkama. Modeli su trenirani na povijesnim podacima, a njihove... More performanse su evaluirane pomoću preciznosti i vizualizirane matricama zabune. Rezultati su pokazali da su svi testirani modeli ostvarili preciznost nešto bolju od slučajnog pogađanja, pri čemu se stablo odluke istaknulo kao generalno najbolji prediktor. Međutim, uočene su razlike u performansama ovisno o specifičnom financijskom instrumentu, što sugerira potrebu za prilagodbom odabira modela svakoj pojedinoj situaciji. Less

Abstract (english)

This thesis explores the application of machine learning techniques, specifically decision trees, random forests, and support vector machines (SVM), in predicting stock and fund price movements. The dataset used encompasses various financial instruments, including stocks of technology companies, energy companies, and international funds. The data was collected from reliable sources, preprocessed, and enriched with additional features. The models were trained on historical data, and... More their performance was evaluated using precision and visualized through confusion matrices. The results showed that all tested models achieved somewhat better precision than random guessing, with the decision tree standing out as the generally best predictor. However, differences in performance were observed depending on the specific financial instrument, suggesting the need for adapting the model selection to each individual situation. Less

• Keywords: strojno učenje; predviđanje kretanja dionica; predviđanje kretanja fondova; stablo odluke; nasumična šuma; strojevi s potpornim vektorima (SVM); financijska analiza; investicijske odluke
• Keywords (english): machine learning; stock price prediction; fund price prediction; decision tree; random forest; support vector machines (SVM); financial analysis; investment decisions
• Language: croatian
• URN:NBN: urn:nbn:hr:168:880007
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2024-10-12 20:50:09