Change search
ReferencesLink to record
Permanent link

Direct link
A comparative study of regression analysis with and without search query data as arepresentation of public opinion
KTH, School of Computer Science and Communication (CSC).
KTH, School of Computer Science and Communication (CSC).
2016 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
En undersökande studie av regressions analys med och utan sökdata som en representation av folkopinion (Swedish)
Abstract [en]

Stock prediction models using search query data is a modern phenomena and arelatively unexplored subject which potentially yields improvements to currentlyestablished prediction algorithms. This thesis will strive to improve an autoregressiveprediction model by analyzing concurrent search query data to conclude whether ornot taking such data into account will improve the prediction model. Multiplealternatives for sources of search query data has been analyzed and Google Trendswas concluded as the most suitable candidate. The thesis found no strong indicatorthat amplifying the autoregressive algorithm with Google Trends data would producebetter stock predictions. There remains to be found an elegant solution to improvingprediction models using Google Trends data.

Place, publisher, year, edition, pages
National Category
Computer Science
URN: urn:nbn:se:kth:diva-187229OAI: diva2:929287
Available from: 2016-05-18 Created: 2016-05-18 Last updated: 2016-05-18Bibliographically approved

Open Access in DiVA

fulltext(1481 kB)16 downloads
File information
File name FULLTEXT01.pdfFile size 1481 kBChecksum SHA-512
Type fulltextMimetype application/pdf

By organisation
School of Computer Science and Communication (CSC)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 16 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

Total: 46 hits
ReferencesLink to record
Permanent link

Direct link