Thesis (M.Sc.) -Cairo University, 2023.

Bibliography: pages 68-73.

Research Question One - Is there a relationship between lagged stock returns
and current trading volume?
Estimated VAR model results show that there is a positive relationship between current
stock market turnover and lagged stock market returns but this relationship is not significant.
This relationship is valid even after accounting for volatility and dispersion as alternative reasons
for high trading volumes (turnover). Further analysis using the granger causality test shows that
return does not granger cause turnover meaning that lagged market returns are not significant in
explaining current market turnover. Graphs from the impulse response functions show that there
is an insignificant positive relationship between lagged stock market returns and current market
turnover.
Generally, the results from the VAR models, results from the granger causality test and results
from the impulse response functions show that there is an insignificant positive relationship
between lagged stock market returns and market turnover
5.2.2 Research Question Two – Do investors on the USE exhibit the overconfidence
bias?
The results from VAR models, the granger causality test and impulse response functions
show that there is an insignificant relationship between lagged stock market returns and market
turnover, it means that the null hypothesis that there is a statistically significant positive
relationship between lagged stock market returns and current market turnover in the Uganda
securities exchange is rejected, therefore, this research does not support the overconfidence
hypothesis that an increase or decrease in lagged stock returns leads to an increase or decrease in
current market turnover. To distinguish the overconfidence effect from the disposition effect, this
research was supposed to carry out tests on both the market wide turnover and individual
security turnover. The results of individual security analysis would help to tell if the observed
overconfidence is not a summation of the disposition effect on the individual security level.
However since the results of the VAR estimates showed that there is no statistically significant
lead-lag relationship between market returns and market turnover, there was no need to carry out
the individual security VAR model analysis. Hence this research concludes that investors on the
Uganda Securities Exchange do not exhibit the overconfidence bias.
5.3 Implications of Findings
This research has added to existing literature in behavioral finance by assessing the
existence of overconfidence bias in a developing stock market. It will be a reference for future
researchers who will intend to expand their understanding of overconfidence bias in developing
stock markets like the Uganda Securities Exchange. It is very important that investors especially
individual investors understand their own ways of how they think, feel and reason. This will
enhance their return on investments, stabilize and improve the efficiency of the stock market.
Investors should also carry out fundamental and technical analysis of stocks before making
64
investment decisions. This will improve their decision making and help them to avoid
psychological and cognitive biases that affect their rationality (Metwally 2023).
The results of this research will help financial advisors to understand the effects of
overconfidence bias on the investment decisions of their clients and hence they can come up with
aggressive strategies aimed at minimizing the negative impacts of this bias. The regulatory body
i.e. the Uganda Capital markets Authority (UCMA) should put into consideration the confidence
level of traders when monitoring the market to plan out new financial policies and strategies
aimed at the efficient operation of the USE. This is because as the number of individual investors
increase, the level of overconfidence will also increase over time and this will affect the volatility
of returns and the general efficiency of the USE.
In order to increase individual investor participation the government through USE and
UCMA should lead in educating the public, especially at individual level about the benefits of
trading on the stock market as an alternative to the traditional investment avenues. To entice
potential investors, the public awareness and education given should address both risks and
benefits so that those wishing to trade will do so from an informed point of view. USE should
fully automate its operating systems just like other stock markets. Embracing ICT would cut
down on the paper documentation process which according to Mshilla et al., (2015) is
considered to be bulky and user-unfriendly.
5.4 research obstacles
The research faced a limitation in accessing recent empirical studies from developing
countries with developing stock markets as most of the recent studies were carried out in
developed countries with developed stock markets. This research also faced a limitation in
accessing data before 2017 as some companies such as uchumi supermarkets delisted from the
exchange before 2017.
Owing to small sample of listed companies on the USE it was not possible to divide the
companies into large capitalization and small capitalization in order to test the existence of
overconfidence on those companies separately
The Uganda securities exchange is going through a digitalization process and so there is no
separate data available on individual and institutional investor accounts. The researcher therefore
failed to compare the manifestation of overconfidence bias in institutional and individual
investors separately. With these limitations in mind, the following recommendations have been
made with the aim of filling the gap or adding to the existing literature in overconfidence bias
and behavioral finance at large
5.5 Future studies
The research recommends considering the disposition effect in future studies. As pointed
out by Statman et al. (2006) overconfidence drives the disposition effect which is the tendency to
sell winning stocks too soon and hold losing stocks for long. Overconfidence and disposition are
65
different in two ways. Firstly the disposition effect is the attitude of an individual towards a stock
in the portfolio but overconfidence affects the whole market. Secondly disposition explains one
side of the trade i.e. the sell side because disposition only happens when investors sell a security
while overconfidence explains both sides of the transaction i.e. buying and selling (Ganesh et al.,
2020). Hence, future studies need to test the overconfidence bias alongside the disposition bias as
it has been put forth that disposition might be another behavioral explanation for trading trends
within the stock market (Metwally et al. 2015)
The research concentrated on testing the existence of overconfidence bias on the USE
only, future studies should test the existence of other behavioral biases such as herding.
Understanding the effects of different behavioral biases on the USE would help solve the
problem of price distortions and improve the efficiency of the stock market
As the USE becomes fully automated, separate data on individual and institutional
investors will be accessible. Future studies should therefore test the presence of overconfidence
bias on the USE using individual investor’s accounts and institutional investor’s accounts
separately. There is a wide range of findings in developed countries showing that individual
investors are more likely to be overconfident (irrational) as compared to institutional investors

النظريات التقليدية في التمويل تفترض أن المستثمر يتخذ قرارات منطقية وأن الأسواق فعّالة. ومع ذلك، فشلت هذه النظريات في تفسير
بعض الظواهر الغريبة الموجودة في سوق الأسهم، على سبيل المثال وجود مستويات عالية من التداول. نقاط الضعف في نظريات التمويل
التقليدية أدت إلى ظهور التمويل السلوكي الذي يعتبر المستثمر كإنسان عادي يتخذ قرارات غير منطقية عند مواجهته لعدم اليقين نتيجة
لتحيزات نفسية مثل تحيز الثقة الزائدة. عدة دراسات سابقة تربط ظاهرة التداول المفرط بتحيز الثقة الزائدة .
قام العديد من الباحثين بفحص وجود تحيز الثقة الزائدة في أسواق الأسهم. ومع ذلك، تمت معظم هذه الدراسات في أسواق الأسهم المتقدمة،
ولقلة الدراسات في الأسواق الناشئة مثل بورصة الأوراق المالية في أوغندا. علاوة على ذلك، كانت بعض هذه الدراسات غير نهائية
والبعض الآخر كان نظريًا يجمع بين الدراسات السابقة. الغرض من هذا البحث هو اختبار بشكل تجريبي ما إذا كان تحيز الثقة الزائدة
موجودًا ببورصة الأوراق المالية في أوغندا. يتم ذلك من خلال فحص العلاقة بين عوائد سوق الأسهم وحجم التداول باستخدام تقنيات
الفحص الإحصائي للسلاسل الزمنية مثل نموذج المتجه ذاتي الانحدار، واختبار السببية ، ودوال الاستجابة الفورية. أظهرت النتائج عدم
وجود علاقة إيجابية غير معنوية بين عوائد سوق الأسهم المتأخرة وحجم التداول الحالي، مما يشير إلى عدم وجود تحيز الثقة الزائدة على
بورصة الأوراق المالية في أوغندا .
كلمات رئيسية: تحيز الثقة الزائدة، بورصة الأوراق المالية في أوغندا ) USE (، عوائد سوق الأسهم، حجم

Issues also as CD.

Text in English and abstract in Arabic & English.