University of Twente Student Theses

Login

Using Sentiment Data from the Global Database for Events, Language and Tone (GDELT) to Predict Short-Term Stock Price Developments

Jakel, Tibor (2019) Using Sentiment Data from the Global Database for Events, Language and Tone (GDELT) to Predict Short-Term Stock Price Developments.

[img] PDF
681kB
Abstract:This study is set to investigate whether media sentiment, retrieved from the Global Database of Events, Language and Tone, can be a predictor of stock price. A business process model for the extraction and utilization of sentiment data from GDELT is presented and used for measuring the cross-correlation between average media sentiment and closing stock price of Facebook, Apple, Amazon, Alphabet and Tesla. A total of more than 5 million news article entries from GDELT were analyzed, 58.655 of which are relevant to this research. Alphabet is the only company with a strong positive cross-correlation of average daily media sentiment with adjusted closing stock price on the same day, while Facebook and Tesla exhibit weak negative correlations.
Item Type:Essay (Bachelor)
Faculty:BMS: Behavioural, Management and Social Sciences
Subject:54 computer science, 58 process technology, 85 business administration, organizational science
Programme:International Business Administration BSc (50952)
Keywords:Big data, Cross-correlation analysis, GDELT, Sentiment analysis, Stock market forecasting
Link to this item:https://purl.utwente.nl/essays/78614
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page