Facebook Sentiment Index and international stock markets

Abu Bakar, Azizah (2017) Facebook Sentiment Index and international stock markets. PhD thesis, University of Glasgow.

Abstract

This thesis aims to provide new behavioural finance insight into market anomalies through the use of a novel approach to measuring investor sentiment: the Facebook’s Gross National Happiness (GNH) index. The three empirical essays of this thesis investigate separately the relation between country-level investor mood – measured daily using the GNH index, and the following occurrences: the Monday Effect, return differentials in cross-listed shares; and herding. The empirical investigations are carried out with data (September 2007 to March 2012) for up to 20 international markets for which daily GNH data are available. In the first essay (Chapter 3), new empirical evidence is provided on the relation between mood and the Monday Effect. This chapter examines whether the well-documented evidence of Monday returns being significantly lower than other trading days of the week relates with mood. The results indicate that the Monday Effect become insignificant when mood is controlled for, and that such effect is more prominent within small capitalization indices and within collectivist and high uncertainty avoidance countries. These findings thus provide empirical support to a behavioural explanation for the Monday Effect, particularly the ‘Blue Monday’ hypothesis. The second essay (Chapter 4) investigates whether returns differentials in pairs of cross-listed shares is related to mood differential between the markets on which these securities are traded. The results, based on 281 pairs of synchronously traded cross-listed shares and controlling for arbitrage costs, support the hypothesized positive relation between return and mood differentials. Such relation is also found to be more prominent among small-cap firms, as these shares tend to attract small investors who are prone to sentiment-induced biases. The third essay (Chapter 5) adds to limited existing empirical evidence on the relation between investor mood and herding behaviour. This chapter investigates whether investors exhibit greater tendency to herd during days of extreme (upper or lower) moods, and whether such relation is affected by firm-size and culture. The results indicate the presence of herding during days with extreme mood, thus lending further empirical support to the notion that herding is mainly driven by psychological factors. Consistent with prior literature, the relation between herding and mood is found to be stronger within small capitalization indices and in countries with collectivist and high uncertainty avoidance culture. Overall, this thesis contributes to further understanding of the effect of investors’ psychological biases on the market.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: culture, cross-listing, facebook, gross national happiness, herding, investor sentiment, monday effect, mood, stock market.
• Subjects: H Social Sciences > HF Commerce > HF5601 Accounting; H Social Sciences > HG Finance
• Colleges/Schools: College of Social Sciences > Adam Smith Business School > Accounting and Finance
• Supervisor's Name: Vagenas-Nanos, Dr. Evangelos and Siganos, Dr. Antonios
• Date of Award: 2017
• Embargo Date: 21 November 2021
• Depositing User: Ms Azizah Abu Bakar
• Unique ID: glathesis:2017-8604
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 28 Nov 2017 10:35
• Last Modified: 23 Nov 2020 05:49
• URI: https://theses.gla.ac.uk/id/eprint/8604