INVESTOR ATTENTION AND SENTIMENT

Wang, Jessica Yichun (2015) INVESTOR ATTENTION AND SENTIMENT. Doctoral thesis, University of East Anglia.

[img]
Preview
PDF
Download (706kB) | Preview

Abstract

Investor sentiment and attention are often linked to the same non-economic events making it
difficult to understand why and how asset prices are affected. This thesis disentangles these
two potential drivers of market behaviour by studying how investors react to sports outcomes,
weather conditions and merger and acquisition announcements.
Firstly, a new dataset of medals for major participating countries and sponsor firms over
four Summer Olympic Games is analysed. Results show that although Olympic success does
not lead to abnormal stock returns, subsequent market activity is reduced substantially. In
the US, for example, trading volume (realised volatility) during Olympics is over 24% (46%)
lower than usual while gold medal awards lead to a further decrease over the next trading day.
These findings are in line with recent theories and evidence related to investor inattention but
cannot easily be explained on the basis of sentiment. Analysis of data from online search
volumes and surveys measuring investor sentiment, also suggest that the market impact of
the Olympics is linked to changes in attention. I demonstrate that the statistical regularities
can be exploited by simple volatility trading strategies in the US to produce significant risk
adjusted profits.
Secondly, I study the relationship between weather and stock market activity using a
new perspective that does not rely solely on investor mood. I argue that bad weather can
increase the productivity of investors by making them more focused on trading and less
concerned about other leisure activities. This allows me to explain the empirical finding
of higher trading activity on rainy days for a sample of 33 international stock markets. In
line with previous literature, I confirm that particularly bad weather conditions which create
inconvenience to market participants, such as snow, have the opposite effect by reducing
productivity and trading volume. Finally, I find evidence that weather has a nonlinear effect
on market activity.
Thirdly, I explore if the market reaction to M&As in the US is governed by attention
or sentiment. I find that attention, as proxied by online abnormal search volume, decreases
significantly before announcements and then increases dramatically on the event date. The
high level of attention diminishes shortly after. I also investigate whether the abnormal search
volume surrounding the event date affects stock prices. The results suggest that the resolved
uncertainty before the announcement date is incorporated into price discovery shortly after
the announcement as the learning capacity of investors constrains the information processing
speed in a bid to adjust the investment decisions.

Item Type: Thesis (Doctoral)
Faculty \ School: Faculty of Social Sciences > Norwich Business School
Depositing User: Vailele Chittock
Date Deposited: 29 Jan 2016 10:41
Last Modified: 29 Jan 2016 10:41
URI: https://ueaeprints.uea.ac.uk/id/eprint/56867
DOI:

Actions (login required)

View Item View Item