This thesis examines how the American economy is affected by the market sentiment that arises from the news about actions and decisions of the American President Donald Trump. The news articles are obtained from Reuters for the period between the 1st of May and the 30th of November 2018, based on which a sentiment variable is created using natural language processing methods. Firstly, the impact of Trump sentiment on the returns on the S&P 500 Index is examined. The results show a positive and statistically significant impact of sentiment from the previous day on today's S&P 500 Index return. A statisti cally significant effect of the sentiment from a week ago is also found, however, this effect is negative. This result indicates that there is an initial overreaction to the new information, followed by subsequent market correction to the mean. Such result is consistent with the findings of the field of behavioural finance, which incorporates the idea that investor psychology is involved in investment decision making. Secondly, the impact of the news sentiment on the performance of individual sectors of the American economy, as measured by the returns on S&P 500 sector indices, is analysed. A statistically significant effect of sentiment on sector index return is found in the case of Consumer...
| Identifer | oai:union.ndltd.org:nusl.cz/oai:invenio.nusl.cz:405661 |
| Date | January 2019 |
| Creators | Pinteková, Aneta |
| Contributors | Kukačka, Jiří, Horváth, Roman |
| Source Sets | Czech ETDs |
| Language | English |
| Detected Language | English |
| Type | info:eu-repo/semantics/masterThesis |
| Rights | info:eu-repo/semantics/restrictedAccess |
Page generated in 0.0041 seconds