kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
AI-baserad förutsägelse av aktierörelser: LSTM och GRU-modeller för den svenska marknaden
KTH, School of Engineering Sciences in Chemistry, Biotechnology and Health (CBH), Biomedical Engineering and Health Systems, Health Informatics and Logistics.
2025 (Swedish)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
AI-Based Prediction of Stock Movements: LSTM and GRU Models for the Swedish Market (English)
Abstract [sv]

Att förutse aktiekursrörelser är en central utmaning för investerare som strävar efteratt minska risker och öka framgången för sina investeringar. Detta examensarbeteundersöker potentialen hos två AI-baserade modeller, Long Short-Term memory(LSTM) och Gated Recurrent Unit (GRU), för att förutsäga kortsiktigaaktiekursrörelser på 5 aktier inom den svenska energisektorn. Modellerna tränadesmed historisk data, inklusive closing price och index, och utvärderades bland annatmed felmåttet Mean Squared Error. Resultaten visade att GRU presterade bättre änLSTM för kortsiktiga prognoser, tack vare dess enkla struktur och effektivareträningsprocess. Studien identifierade även indexinkludering som både har positivoch negativ effekt på resultatet beroende på modell och tidsperiod. Arbetet bidrar tillatt förbättra förutsägelser av aktier och föreslår flera områden för framtida forskning,såsom integration av fler parametrar och avancerade AI-modeller. 

Abstract [en]

Predicting stock price movements is a central challenge for investors seeking toreduce risks and increase the success of their investments. This thesis examines thepotential of two AI-based models, Long Short-Term Memory (LSTM) and GatedRecurrent Unit (GRU), to predict short-term stock price movements for five stocksin the Swedish energy sector. The models were trained on historical data, includingclosing prices and indices, and were evaluated using metrics such as Mean SquaredError. The results showed that GRU outperformed LSTM for short-term forecastsdue to its simpler structure and more efficient training process. The study alsoidentified that the inclusion of indices has both positive and negative effects on theresult, depending on the model and the time period. This work contributes toimproving stock predictions and suggests several areas for future research, such asthe integration of additional parameters and advanced AI models. 

Place, publisher, year, edition, pages
2025.
Series
TRITA-CBH-GRU ; 006
Keywords [en]
Machine learning, artificiall intelligence, RNN, LSTM, GRU, stock market, index, closing price, mean squared error
Keywords [sv]
Maskininlärning, artificiell intelligens, RNN, LSTM, GRU, aktiemarknad, index, closing price, mean squared error
National Category
Artificial Intelligence
Identifiers
URN: urn:nbn:se:kth:diva-360009OAI: oai:DiVA.org:kth-360009DiVA, id: diva2:1937627
External cooperation
Zenon AB
Subject / course
Computer Engineering with Business Economics
Educational program
Bachelor of Science in Engineering - Computer Engineering and Economics
Supervisors
Examiners
Available from: 2025-02-14 Created: 2025-02-13 Last updated: 2025-02-14Bibliographically approved

Open Access in DiVA

Examensarbete Sannaz(4115 kB)132 downloads
File information
File name FULLTEXT01.pdfFile size 4115 kBChecksum SHA-512
ece4d738b80bbb40eb879be2601c0f9b5a2e21267048a34c36c49d8bab4ddbba7a842c7b90ecbfc603d422df0ba5c6900b0e208be2b9185a9bcee7ce69f0a00c
Type fulltextMimetype application/pdf

By organisation
Health Informatics and Logistics
Artificial Intelligence

Search outside of DiVA

GoogleGoogle Scholar
Total: 132 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 702 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf