Presidential rhetoric, sentiment and their relation to stock markets
| Thesis title in Czech: | Presidential rhetoric, sentiment and their relation to stock markets |
|---|---|
| Thesis title in English: | Presidential rhetoric, sentiment and their relation to stock markets |
| English key words: | sentiment analysis, stock markets, Granger causality, presidential rhetoric, natural language processing |
| Academic year of topic announcement: | 2016/2017 |
| Thesis type: | Bachelor's thesis |
| Thesis language: | angličtina |
| Department: | Institute of Economic Studies (23-IES) |
| Supervisor: | PhDr. Boril Šopov, M.Sc., LL.M. |
| Author: | hidden - assigned by the advisor |
| Date of registration: | 09.11.2016 |
| Date of assignment: | 09.11.2016 |
| Date and time of defence: | 13.06.2017 09:00 |
| Venue of defence: | Opletalova - Opletalova 26, O105, Opletalova - místn. č. 105 |
| Date of electronic submission: | 19.05.2017 |
| Date of proceeded defence: | 13.06.2017 |
| Opponents: | PhDr. Diana Žigraiová, Ph.D. |
| URKUND check: | |
| Guidelines |
| The bivariate dependence modelling will be used, thus the dependencies in the pairs of stock prices in various stock markets will be analysed within specific regions. The paper would build on Extreme value theory, which is focused on dependence in extreme values. |
| References |
| 1. Bird, Steven, Ewan Klein, and Edward Loper. Natural language
processing with Python: analyzing text with the natural language toolkit. O’Reilly Media, Inc., 2009. 2. Hutto, Clayton J., and Eric Gilbert. Vader: A parsimonious rulebased model for sentiment analysis of social media text. Eighth international AAAI conference on weblogs and social media. 2014. 3. Bollen, Johan, Huina Mao, and Xiaojun Zeng. Twitter mood predicts the stock market. Journal of computational science 2.1 (2011): 1-8. 4. Mittal, Anshul, and Arpit Goel. Stock prediction using twitter sentiment analysis. Stanford University, CS229 (http://cs229.stanford.e du/proj2011/GoelMittal-StockMarketPredictionUsingTwitterSent imentAnalysis.pdf) 15 (2012). 5. Wooldridge, Jeffrey M. Introductory econometrics: A modern approach. Nelson Education, 2015. |
| Preliminary scope of work in English |
| A vast number of researchers have been concerned with the question regarding
the factors that have predictive power on stock markets. Numerous studies emerged over the last decades trying to relate the stock market‘s movement to sentiment extracted from the big data. The focus of their research have been aimed mainly at the public mood values hidden within the textual content posted on social networks, such as microblogging website Twitter. Moreover, the presidential rhetoric on the social networks has received a lot of the public attention recently, and especially the Twitter feed of the incumbent President of the United States, Donald Trump, has appeared uncountable times in the headlines of the newspapers all over the world. With the degree of power that is so characteristic for every President of the United States, it might be worth studying if there is a relation of his remarks posted online and the stock indexes‘ movements. Various studies have examined the rhetoric of the President of the US and its impact on economy, but none of them has been concerned with his Twitter feed and its relation to stock markets so far. Thus, the goal of this thesis is to fill this gap and examine the Twitter time lines of two consecutive US Presidents, 7 Barack Obama and Donald Trump. Not only is the study concerned with the textual analysis and comparison of the characteristics of their online rhetoric using Natural Language Processing techniques, but also tries to relate the emotional states extracted from their tweets to the selected stock indexes. Methodology: Firstly, the VADER, lexicon – based model for sentiment analysis would be employed on the Twitter data sets over the course of their presidential mandate, for which the data is available of two consecutive US presidents, Barack Obama (@BarackObama) and Donald Trump (@realDonaldTrump). The Granger causality analysis of the compound sentiment time series aggregated into trading days in the data sets and three stock indexes, namely DJIA, S&P 500, and NASDAQ would be carried out consequently. Hypotheses: 1. The sentiment extracted from tweets posted by Donald Trump Granger causes the time series of selected stock indexes. 2. The sentiment extracted from tweets posted by Barack Obama Granger causes the time series of selected stock indexes. |