This thesis examines issues related to initial public offering (IPO) and venture capital activities. The thesis first investigates venture capitalists' exits (Le. through IPO, M&A, Liquidation and LBO routes) from their investee companies for a sample of 5059 investments by UK venture capital firms during the period between 1990 and 2006. The time to exit is modelled non-parametrically using cumulative distribution function and parametrically using a frailty model and cumulative distribution function. The evidence shows that venture capital investors tend to exit their portfolios within 7 years after the investments. The time to exit from portfolio companies are shorter in North America than in Europe and the rest of the world. This is likely to be the effect of a developed venture capital market in North America more than in Europe and the rest of the world. I find that experienced venture capital firms as measured by age tend to hold their investments in portfolio companies much longer than inexperienced venture capital providers. In addition, venture capital investors are likely to exit mature portfolios through the M&A route, while young target companies do so through the IPO route. The thesis also evaluates the risk and returns of venture capital syndicated investments between the US and Europe from 1995 to 2006 for a sample of 8780 investments. Of these 6008 investments are exited through IPO route and 2780 are exited via non-IPO method. I find that venture capital returns are higher than market returns. For early stage investments the average returns are 111 percent, while for expansion and later stage investments the returns are 97 and 82.2 percent respectively. I find evidence that the returns are high during the bubble period and also for some sectors (Le. Information technology, Noncyclical consumer goods and Non-cyclical service). The systematic risk (beta) is higher than the market beta for early stage investments. Third and final, the thesis investigates the survival profiles of companies across the globe floated on the Alternative Investments Market (AIM) from its inception in 1995 to the end of 2004. In addition, it compares the survival rates of venture and non-venture backed companies. For a sample of 918 companies admitted to the market, I find survival rates are broadly comparable to North American IPOs, and conclude that, contrary to allegation by the US stock market regulator, AIM is not a casino. I identify four regulatory levers: (minimum) requirements on public float, company age (or trading record), size (Le., market capitalization) and the role (reputation) of the nominated advisors (Nomad). I find that the chances of survival increase with three regulatory levers (company age, size, and Nomad reputation). I find that IPOs incorporated in the UK have higher survival rates than non-UK incorporated IPOs by approximately 2.29years. I also find that consistent with US evidence, initial IPO returns have a positive impact on survival rate.

No description available.

No description available.

Modern Finance literature persistently ignores the systemically destabilizing effects of financial bubbles. As a result, periodic speculative excesses, which hugely deviate from the rational models of mainstream finance, are largely unexplored, especially with regard to institutional investors’ behaviour in financially euphoric environments. My main objectives are to expose the premises, which the speculative bubble was built on, and the factors affecting institutional investors’ investment decisions, objectives and risk attitude in speculative bubbles. Using a series of semi-structured interviews with fund managers that worked during the Cyprus bubble of 1999, this thesis aims to contribute to the limited literature regarding institutional investors’ speculation. I draw from Abolafia and Kilduff, Kindleberger, Minsky, and Galbraith in order to provide a descriptive framework of speculative bubbles, in which institutional investors appear to be purposive, contrary to and at the expense of retail investors and the systemic stability. The empirical data suggest that the roots of speculative bubbles are set by an event with perceived real economic consequences, which is seen to improve economic conditions and shift investors’ expectations. Afterwards, the rising share prices keep inviting an increasing number of speculators who create a new reality by replacing reason with what appears to be misinterpretation and misunderstanding. In this environment, regulatory failure, rumours and ‘strange friendships’ appear on the scene. Additionally, there is strong evidence suggesting that the institutional investors’ understanding of risk in speculative markets, contrary to the conventional wisdom, is particularly problematic; a phenomenon I call ‘risk paradox’. The implications of speculative bubbles and institutional investors’ risk attitude are crucial in understanding the limitations of rational models that prevail in finance. This thesis argues for situating investment activity within its social, and frequently, speculative context. It contributes to understanding the behaviour of institutional investors in speculative markets and calls attention to their irrational investment behaviour.

This thesis contains 3 empirical chapters with relevance to the ‘international stock market integration’ literature. The first chapter aims to investigate the evolution of the international integration of emerging European stock markets with the world market for the period of 1996 to 2011. For this purpose, using dynamic conditional correlation approaches, the changes that occur in correlation (integration) levels due to four global/regional incidents: i) the 1998 Russian crisis; ii) the 2001 dotcom crisis and the 9/11 shocks; iii) the 2004 EU enlargement; and iv) the 2007-2009 global financial crisis are examined. The findings show that the volatilities of emerging European stock markets and their correlation structures with the world market significantly change due to the impacts of global/regional incidents. Although it is obvious that each incident has a differential impact on each country depending on the internal dynamics of those countries at the times of incidents, the findings still clearly reveal the general common impacts of the investigated incidents on the volatilities and the correlation structures of the sample countries with the world market. The second chapter investigates the international stock market integration phenomenon at a disaggregated level for emerging European countries. For this purpose, by using the Geweke technique (1982) the world market integration levels of individual companies, namely ‘individual stock integrations’ are measured. Furthermore, by using firm specific and industry level variables, the year to year changes in integration levels are explained to identify the determinants of an individual level stock integration. The results confirm the presence of individual stock integration since each company is integrated with the world market at different level of strength. Furthermore, panel data analysis shows that it is possible to explain those differences on the individual integration levels with both company specific variables and industry level variables. Comparing tranquil and shock periods’ heteroscedasticity corrected conditional correlations and dynamic conditional correlations; the final chapter tests the widely accepted belief of the significance of a ‘contagion effect’ from the US to emerging European countries during the latest global financial crisis. The chapter reveals that although the contagion effect is the most blamed factor for the propagation of financial crises, particularly for the last global financial crisis, the presence of contagion effect from the US market (crisis-origin country) is not that certain since the conclusion is highly dependent on econometric specifications and sample period diversifications.

The thesis investigates convergence between UK, US and European stock markets (1974-2001), to address four main questions: Are movements in the market rise premium for the UK associated with movements in the market risk premium for the US, or are they associated with movements in the market risk premium for European countries; are stock market prices converging over long horizons and what may drive such convergence; are the characteristics of the market risk premium similar to the characteristics of value risk premium; and, finally, can the value premium be linked to economic business conditions in such a way that convergence of this risk factor across national boundaries is possible? The main conclusions from this thesis are (i) The UK market risk premium appears to be converging with its US counterpart rather than market risk premium in European stock markets. (ii) A common fundamental factor is more pronounced between UK and US stock markets rather that than the European markets. (iii) The evidence supporting the convergence of market risk premium between the UK and the US is not replicated for the value premium, where convergence trends are much less marked. (iv) The evidence on the source of the value premium suggests that the rational explanation and subsequent link to business conditions is relevant for the US stock market only. Subsequently, it is perhaps unsurprising to find that there is little evidence of global convergence for the value premium. Overall, these results imply that the time may not be right for any acceleration of merger activity between the UK stock market and its European neighbours.

Research studies (such as Kiyilar and Acar, 2009) have supported that investors act in irrational ways in some of their investment decisions, and financial models have failed to explain the real investors' behaviour. Investors' trading activity is influenced by personality traits and psychological biases (overconfidence, risk tolerance, self-monitoring, and social influence) and is also affected by mood. The aim of the thesis is to confirm these assumptions by developing and testing a model (using SEM analysis) which would incorporate and examine all of them simultaneously, as it actually happens in real life. The research population includes 345 Greek investors, including individuals' as w~rt' 'as professional' investors who work in various investment companies located all over the country. The data collection included two surveys. The first survey investigated psychological biases and personality traits to find if they correlate with stock trading performance, whereas the second survey examined the psychological predisposition to find whether mood affects stock trading performance. The results have verified that these psychological biases, personality traits and mood, influence investors' trading performance, frequency and volume, providing a complete research model. Another objective of this study was to understand the profile of Greek investors and test if there are differences among them as far as stock trading behaviour (performance, volume and frequency) is concerned. Cluster analysis (three-cluster solution) identified three investors' profiles, the low, moderate and high investor profile, and revealed that high profile investors (with the higher scores on the psychological biases and personality traits) trade high volumes of stocks, make transactions more frequently and earn higher stock profits compared to investors belonging to the other two profiles. A comparative analysis between professionals and individuals has shown that professional investors have higher performance than individuals as far as stock trading is concerned. The results have also shown that professional investors score high on the psychological biases and personality traits examined. The second stage of the study has required iterative data using questions that depict psychological predisposition in a dynamic way. The cluster analysis of 1 Non-professionals 2 Portfolio analysts and stockbrokers 1 the second data set has identified and compared different mood status highlighting differences among groups as far as their mood status and their stock trading performance is concerned. This study has provided evidence supporting the significance of some subjective factors, such as personality traits, psychological biases and emotions in investors' trading activity. The findings have shown that investors who have specific characteristics such as overconfidence, high self-monitoring, risk intolerance, positive mood and sociability are probably better on stock trading performance. ~'A This thesis could enable individual investors and inv;stment advisors, to construct a framework of the profile that contributes to high stock trading performance (a stock trading performance guide). Therefore, investors could possibly ensure the stock trading performance, to the extent that this depends on their profile. Moreover, the study contributes to the field providing a complete and verified research model concerning investors' trading behaviour. Additionally, a contribution of this study is the extensive literature review in the field of behavioural finance which provides a better understanding of behavioural factors and a framework for academics, researchers, individual and professional investors. Keywords: Behavioural Finance; Trading Behaviour; Trading Activity; Model Analysis. JEL Classification: C30, D14, Gll, 016

This thesis considers the short term stock market reaction to the launch of R&D projects, the implications to the risk characteristics of the announcing firms, and the long term market valuation ofR&D active firms in the UK's stock exchange. As part of the work a survey of the literature on the valuation of R&D investments and expenditures is carried out. The results show that in the short-term, the market reacts positively to R&D project announcements although the magnitude of the reaction is larger for firms with high R&D intensity. However, contrary to the predictions of real options theory, there is no evidence that these announcements result in a change in risk characteristics. Meanwhile, in relation to the long term valuation of R&D in the UK's stock markets, the results show that the market does account for R&D expenditures over the period 1996 to 2006. However, there is still uncertainty around R&D active stocks. The overall aim of this work is to contribute to the literature on R&D in the UK and provide empirical evidence with the small data that is available and the short cycle that technology stocks have experienced.

This thesis explores the link between earnings management and forecast accuracy in the context of Malaysian IPO's following a revision of the regulation on earnings forecast disclosure made in 1996. The study involves three different stages. The first stage examines the accuracy of earnings forecasts contained in the IPO prospectuses of Malaysian companies seeking listing from 1996 to December 2002. The second stage of study provides evidence of positive discretionary accrual in financial statements of IPO issuers in the year of IPO, and in the 3 year period following the IPO. Finally, a correlation study examines the link between earnings management and forecast error and other variables representing unexpected change in economic condition and company specific characteristics. The results from the first stage of study indicate that Malaysian IPO companies on average have a negative forecast error, indicating positive bias in their forecast. Multivariate results indicate that regulation of earnings forecast disclosure has no significant impact on accuracy but that economic condition, management optimism, and auditor reputation have. The second stage, studying earnings management on a sample of IPO 1996,1998 and 2000 regulated companies, provides evidence consistent with the prediction that managers of Malaysian IPO companies manage earnings upwards in the year of forecast issuance, or in the year the company make their forecasts. The study also provides evidence that managers continue to manage earnings during the period after listing, so long as there is continuing regulatory scrutiny. The findings of the final stage of study provide evidence of a significant association between earnings management and the relative size and direction of forecast error, after controlling for other expected associations. The regression results reveal that earnings management of Malaysian IPO companies is associated with forecast error, the changes in economic condition represented by a recovery and crisis period, company age and management ownership. The study makes a contribution in terms of understanding the nature of earnings management at the time of an IPO and in particular providing empirical evidence on the link between the forecast error and the extent of earnings management. The result shows that managers appear to manage earnings upwards significantly during the economic crisis and recovery period in order to match or come closer to the forecast made in the prospectus. In a highly concentrated ownership, the actions of IPO managers appear to be contrary to the assumption of agency theory. It is speculated that managers of IPO companies are managing their earnings upwards and reporting towards meeting their forecasts in order to manage their legitimacy and to establish their company's good reputation. This is because, as newly listed companies, they are under close market scrutiny and are under great pressure to meet the projections made to investors.

This thesis is a comprehensive study how information explicitness - the quantity of detail in a piece of information - affects investors behaviour in stock markets, from the perspectives of experimental and real-stork-market contexts. The core question addressed by this thesis , whether investors always benefit from analysing more dctailed l information in a stock market . ll1lhe first stage. surveys are conducted among unsophisticated (students) and sophisticated (financial industry staff) investors, who arc required \0 for'1:cast the probability of occurrence of uncertain events and make trading decisions accordingly. The results generated from the two surveys indicate that both sophisticated and unsophisticated investors are prone to project a higher probability of occurrence when an uncertain main event is informed by more detailed (explicit) information, However. the impact of information explicitness on investors trading decisions is not significant. Moving toward a situation in a real stock market. a trading experiment based on a trading simulations carried out in a laboratory. . As well as confirming that explicit information increases investors' judged probabilities of occurrence On uncertain events, the results also extend the conclusions regarding trading decisions and performance to show that information explicitness significantly drives investors' performance efficiency (shortening their reaction times but impairing the accuracy of their activities) but not their performance effectiveness (return, trading volumes, and holding positions)_ Finally, an empirical study focuses on how abnormal sentiment information released by FTSE 100 companies and by newspapers can affect market performance. . Although the results are consistent with previous studies, which have shown that negative information inversely affects returns and positively influence~ trading volumes, this study also convey ' that the impact of negative information is restricted to the listed companies' original disclosure and is mainly observed on the announcement day. Additionally, there is an interaction between the abnormal occurrence of negative words and media coverage. Newspaper stories dilute the effect of negative words in companies announcements on returns, while strengthening their impact on trading volumes.