Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
News, sentiment and the real economy
Thesis (PhD)--Stellenbosch University, 2020.
Saved in:
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | en_ZA |
| Published: |
Stellenbosch : Stellenbosch University
2020
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| _version_ | 1867613918292082688 |
|---|---|
| access_status_str | Open Access |
| author | Odendaal, Hanjo |
| author2 | Kirsten, Johann |
| author_browse | Kirsten, Johann Odendaal, Hanjo |
| author_facet | Kirsten, Johann Odendaal, Hanjo |
| author_sort | Odendaal, Hanjo |
| collection | Thesis |
| dc_rights_str_mv | Stellenbosch University |
| description | Thesis (PhD)--Stellenbosch University, 2020. |
| format | Thesis |
| id | oai:scholar.sun.ac.za:10019.1/109134 |
| institution | Stellenbosch University (South Africa) |
| language | en_ZA |
| last_indexed | 2026-06-10T12:43:46.817Z |
| license_str | Other — see source repository |
| provenance_str_mv | Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository |
| publishDate | 2020 |
| publishDateRange | 2020 |
| publishDateSort | 2020 |
| publisher | Stellenbosch : Stellenbosch University |
| publisherStr | Stellenbosch : Stellenbosch University |
| record_format | dspace |
| source_str | SUNScholar — Stellenbosch University Repository |
| spelling | oai:scholar.sun.ac.za:10019.1/109134 News, sentiment and the real economy Odendaal, Hanjo Kirsten, Johann Reid, Monique B. Stellenbosch University. Faculty of Economic and Management Sciences. Dept. of Economics. Big data Text data mining Sentometrics Sentiment analysis Consumer confidence UCTD Thesis (PhD)--Stellenbosch University, 2020. ENGLISH SUMMARY : In this dissertation, text analysis is presented as a complement to traditional survey-based methods used to capture sentiment. This is achieved by firstly constructing media-based sentiment indices from a large variety of news sources for South Africa and presenting these indices as a feasible way to replicate the results of the traditional survey-based consumer confidence index (CCI). The findings of the cointegration and Granger-causality tests support the hypothesis that news-based indices could possibly be used to address shortcomings commonly experienced in the survey-based alternative. The second contribution towards the literature is the evaluation of the adequacy of media-based indices as a predictor of personal consumption. The predictive power of media sentiment indices are evaluated in a Bayesian forecasting horse race alongside the CCI. The conclusion revealed that the inclusion of media-based sentiment indices as predictors in a model can decrease forecasting errors of personal consumption expenditure. The forecasting errors decreased in the cases of both short and long (up to 2 years) forecasting horizons. The results substantiate the theory that news media sentiment contains information on the coincidental and future state of the economy which is not captured in the CCI. The results suggest that media based indices could function as both a complement or alternative to the CCI in consumption forecasting. The final contribution of the thesis showcases the effectiveness of utilizing domain-specific dictionaries to capture sentiment. In the last chapter, domain-specific dictionaries are constructed, in an automated fashion, using Random Forests. These domain-specific indices successfully capture economic sentiment more accurately than the widely used Loughran dictionary. This framework reduces the resources required to extract information from media reports into a sentiment dictionary, while also maintaining a level of transparency. Collectively, the results presented in this dissertation offer some initial support for the use of text analysis in South Africa as an alternative way of capturing softer economic indicators such as economic sentiment. AFRIKAANSE OPSOMMING : Hierdie tesis het dit ten doel om te bewys dat teksanalise 'n aanvullende rol kan bied vir tradisionele opname-gebaseerde metodes, as 'n manier om sentiment vas te lê. Dit word gedoen deur media-gebaseerde sentiment-indekse uit 'n groot verskeidenheid nuusbronne in Suid-Afrika op te stel en hierdie indekse voor te stel as uitvoerbare duplikate van die tradisionele indeks vir verbruikersvertroue indeks (VVI). Die bevindinge van die co-integrasie en Granger-causality toetse ondersteun die hipotese dat nuusgebaseerde indekse gebruik kan word om die tekortkominge in die huidige opname aan te spreek. Die tweede bydrae tot die literatuur is die evaluering van die uitvoerbaarheid van mediagebaseerde indekse, as 'n aanvulling of 'n vervanging van die VVI, as voorspellers van persoonlike verbruik. Die vooruitskattingsvermo ë van mediasentiment-indekse word beoordeel in 'n voorspellingswedren met die VVI. Die resultate van die voorspellingsoefening het aan die lig gebring dat die insluiting van media-gebaseerde sentiment-indekse as voorspellers in 'n model die voorspellingsfoute vir persoonlike verbruiksbesteding kan verminder. Die voorspellingsfoute het verminder in beide die korttermyn en langtermyn (tot en met twee jaar). Die uitslae bevestig die teorie dat sentiment in die nuusmedia inligting bevat oor die toevallige en vooruitskouende toestand van die ekonomie wat nie in die VVI vasgelê is nie. Die laaste bydrae beklemtoon die doeltreffendheid van die gebruik van domeinspesifieke woordeboeke wanneer analiste probeer om sentiment vas te lê. Domeinspesifieke woordeboek is outomaties opgestel deur gebruik te maak van masjienleertegnieke. Hierdie domeinspesifieke indekse vang die ekonomiese sentiment, wat deur verskillende algemeen gerapporteerde vertrouensindekse voorgestel word, meer akkuraat vas as wat die tradisioneel gebruikte Loughran-woordeboek kan. Hierdie raamwerk vergemaklik die proses van subjektiewe inligting-onttrekking uit media in 'n sentimenteboek, en handhaaf ook 'n vlak van deursigtigheid waartoe woorde vervat is in die konstruksie van die sentiment-indeks. Gesamentlik bied die resultate wat in hierdie proefskrif aangebied word, aanvanklike ondersteuning vir die gebruik van teksanalise in Suid-Afrika as 'n alternatiewe manier om sagter ekonomiese aanwysers soos ekonomiese sentiment vas te lê. Doctoral 2020-11-02T09:05:05Z 2021-01-31T19:36:36Z 2020-11-02T09:05:05Z 2021-01-31T19:36:36Z 2020-12 Thesis http://hdl.handle.net/10019.1/109134 en_ZA Stellenbosch University xvi, 178 pages ; illustrations, includes annexures application/pdf Stellenbosch : Stellenbosch University |
| spellingShingle | Big data Text data mining Sentometrics Sentiment analysis Consumer confidence UCTD Odendaal, Hanjo News, sentiment and the real economy |
| title | News, sentiment and the real economy |
| title_full | News, sentiment and the real economy |
| title_fullStr | News, sentiment and the real economy |
| title_full_unstemmed | News, sentiment and the real economy |
| title_short | News, sentiment and the real economy |
| title_sort | news sentiment and the real economy |
| topic | Big data Text data mining Sentometrics Sentiment analysis Consumer confidence UCTD |
| url | http://hdl.handle.net/10019.1/109134 |
| work_keys_str_mv | AT odendaalhanjo newssentimentandtherealeconomy |