本文主要在探討在何種情況下(意即金融危機發生前後)會發生比較顯著的跨國從眾現象。本文採用2003年10月1日到2009年2月28日期間的台灣加權指數、台灣50、美國S&P 500和道瓊工業指數報酬率資料，並利用Hwang and Salmon (2004)測量從眾行為指標的方法，檢驗在此段期間下，台灣投資人行為是否有明顯跟隨美國投資人行為的現象。實證研究發現整體而言在此段期間下，台灣存在顯著的跨國從眾行為。細部以月為單位探討從眾行為顯著結果的變化，可以歸納出以下三個結論：一、從眾行為主要發生在經濟情況相對穩定的情況下，意即在金融危機日趨嚴重以前，測量從眾行為的指標反而比較顯著。例如在2007年和2008年時的經濟情況比在2003年、2004年和2005年時還要衰退，但測量到存在顯著跨國從眾行為的月數反而較少。二、持續存在顯著跨國從眾行為的最長期間為2005年3月到2006年1月，歷經11個月。三、最常被觀測到有顯著從眾行為現象的月份為1月、11月和12月。 / This paper mainly examines under what conditions herding behavior is likely to become more significant and obvious, in which I modify Hwang and Salmon (2004)’s methodology and use the returns data of Taiwan Weighted Index, Taiwan 50 stock Index, S&P 500 stock and Dow Jones Industry Index of the sample period of October 1, 2003 to February 28, 2009 to test if there is any multinational herding behavior. I find that Taiwan investors in this sample period follow (even imitate) the investment actions of American investors. In more details about the herding patterns, we have found three main phenomena. First, herding behavior mainly occurs significantly during relatively quiet period, say, before the financial crisis, rather than when the market is under stress. The economic situations in 2007 and 2008 are much worse than in 2003, 2004, and 2005, but numbers of months exhibiting significant herding in these bad situations are less. Second, the longest lasting time of herding is March 2005 to January 2006, which totally lasts for eleven months, and this period is before 2007 and 2008 in which the financial systems are destroyed badly. Third, herding always happens in January, November and December given the sample period.

Submitted by Johnny Rodrigues (johnnyrodrigues@ufcg.edu.br) on 2018-06-12T13:47:55Z No. of bitstreams: 1 JOSÉ GILDO DE ARAÚJO JÚNIOR - DISSERTAÇÃO PPGCC 2016..pdf: 44985365 bytes, checksum: 7cf3e353444964c334025b7c6f4f6df5 (MD5) / Made available in DSpace on 2018-06-12T13:47:55Z (GMT). No. of bitstreams: 1 JOSÉ GILDO DE ARAÚJO JÚNIOR - DISSERTAÇÃO PPGCC 2016..pdf: 44985365 bytes, checksum: 7cf3e353444964c334025b7c6f4f6df5 (MD5) Previous issue date: 2016-12-13 / CNPq / Há algum tempo pesquisadores e analistas de mercado vêm apresentando indícios da previsibilidade de mercados acionários. Embora acredite-se que o mercado de ações seja imprevisível, análises de previsibilidade realizadas em bolsas da China, Turquia, Hong Kong, Itália, Teerã e EUA vêm mostrando o contrário. O fato é que a hipótese de eficiência de mercado foi planteada em 1970, e não se poderia prever as mudanças culturais e tecnológicas que impactaram o mundo, como o aumento da capacidade de processamento dos computadores, o desenvolvimento de técnicas de aprendizagem de máquina, a publicação de notícias online e a exposição em tempo real da opinião de investidores em redes sociais, por exemplo. A combinação destes elementos passaram a potencializar o lucro de investidores à medida que simplificaram o monitoramento e a gestão do risco, a compreensão do cenário econômico e até a realização de análises complexas sobre setores, índices e ações em poucos minutos. Este trabalho se propôs a lançar luz sobre relações e impactos que as notícias econômicas publicadas em jornais brasileiros, online, mantêm com o mercado acionário nacional em dois níveis de análise: índice Bovespa e setores. Inicialmente, foram coletadas notícias econômicas publicadas em jornais de alta circulação no Brasil entre 2000 e 2015, seus comentários e suas repercussões nas redes sociais Twitter, Facebook, Linkedln e GooglePlus. A análise de correlação entre o índice Bovespa e a quantidade de compartilhamento de notícias em redes sociais revelam uma correlação negativa de 48%. Além disso, por meio da análise de sentimento das notícias coletadas, verificou-se que a quantidade de notícias positivas publicadas é, em média, 4.5 vezes superior ao de negativas, e que, apesar disso, as notícias negativas são mais repercutidas nas redes sociais que as positivas. Para os setores, verificou-se que o setor mais previsível apenas por meio de notícias econômicas é o setor de Petróleo, Gás e Biocombustíveis enquanto o menos previsível é o setor Bens Industriais. Por fim, as variáveis extraídas das notícias foram utilizadas como base no desenvolvimento de modelos de predição tanto para o índice Bovespa quanto para os setores da BM&FBOVESPA. De forma geral, os resultados encontrados superaram estatisticamente baselines comumente utilizados em ~ 20%. / For some time researchers and market analysts have shown evidence of predictability of stock markets. Although many investors believe that the stock market is unpredictable, predictability analysis in China, Turkey, Hong Kong, Italy, Tehran and the US stock markets has shown the opposite situation. The Efficient-Market Hypothesis (EMH) was designed in 1970 and could not anticipate the cultural and technological changes that affected the world, such as the increased processing power of computers, the development of machine learning techniques, real time publication of news and opinions of investors in social media platforms, such as twitter and facebook, for example. The combination of these elements enabled investors to perform more complex analysis of sectors, índices and stocks in almost real time, thus increasing their understanding of the stock market dynamics and improving their likelihood of success. his study aimed to shed light on the relationships and impacts that economic news published in online Brazilian newspapers, have with the national stock market in two leveis of analysis: Bovespa Index and sectors. Initially, we collected economic news published in high-circulation newspapers in Brazil between 2000 and 2015, their comments and their repercussions on social medias like Twitter, Facebook, Linkedln and GooglePlus. The correlation analysis between the Bovespa index and the amount of news shared on social networks showed a negative correlation of 48%. Furthermore, using sentrment analysis it was found that the amount of positive news reported is in average of 4.5 times higher than the negative, and, nonetheless, the negative news are more rebound on the social media than positive news. For the sectors, it was found that the most predictable sector by economic news is the Oil, Gas and Biofuels while the less predictable is the Industrial Goods sector. Finally, the variables drawn from the news were used as as input for the definition of prediction models for both the Bovespa Index and for the sectors of BM& FBOVESPA. In general, the results overperformed baselines such as the random classifier in ~ 20%.

The financial crisisand the European debt crisis wreaked havoc on many European economies and stock markets. Previous studies have shown that crises are associated with high debt and linked with lower growth. Studies also suggest that politicians underestimate the risks associated with high debt during economic upturn and that economic crises are usually connected with high volatility. Volatility is used as a measurement of risk since high volatility indicates larger market uncertainty of the valuation of the underlying asset. Previous studies have shown that volatility can be a good indication of a firm’s riskiness. As volatility and capital structure both relate to risk and are influenced by market reactions, investigating the impact that capital structure has on volatility during times of global financial market distress could provide insight and be an important tool for investors. This thesis will investigate firms listed on the Stockholm stock exchange, divided into seven industries, in order to find the impact capital structure may have on volatility, before, during and in the aftermath of the recent European debt crisis (2006-2016). The study will use a quantitative research method, with an objectivistic and positivistic research philosophy as well as a deductive research approach. By using multiple regression models, theoretical relations surrounding volatility and capital structure will be contrasted to the results of the study.The results of the study finds that capital structure does not play a significant part inchanges in volatility for firms during any investigated period when testing for all firms simultaneously. However, the same claim cannot be made for when each industry is tested individually. Empirical evidence showed that capital structure is a influencer for changes in volatility for the consumer goods industry prior to and after the debt crisis and in the consumer goods service industry after the debt crisis. Investors are urged to not be concerned by large debt levels, as long as they invest in largefirms and choose the right sectors. The financial sector is seen as the least risky, with low volatility levels. Furthermore, the basic material sector, despite outward appearances, should be avoided as it presents recent periods of unusually large volatility levels.

The work deals with an automatic trading system and adaptive training. Is used both technical and automatic fundamental analyses, therefore as inputs to the neural network is used historical data exchanges and text data from reports. It explains the basics of trading, technical analysis and technical terms. The work deals with technical and fundamental analysis. It contains a description of algorithmic nature, program implementation and experiment with developed trading system. The selected strategy is compared to other approaches.

Bakgrund: En av de vanligaste aspekterna vid investeringar är att åstadkomma hög avkastning samtidigt som risken hålls på en låg nivå. Risker kopplade till bolag beror på flera olika faktorer varaven anses vara bolagsstorleken. För att vidare nå önskad avkastning har flertalet investeringsstrategier presenterats som sägs ska kunna överavkasta marknaden och vilket kontrasterar den effektiva marknadshypotesen. Genom investeringar i bolag med låga börsvärden, samt i kombination medinvesteringsstrategier, är det av intresse att undersöka huruvida överträffande av marknaden kan ske. Syfte: Denna uppsats har som syfte att analysera om det, på Spotlight Stock Market och First North, under åren 2007–2019, har gått att påvisa existensen av en SBE och huruvida det går att generera en riskjusterad överavkastning i småbolag med hjälp av investeringsstrategier. Metod: För att genomföra denna uppsats användes en kvantitativ design i kombination med endeduktiv ansats. Åtta olika portföljer sammanställdes baserat på storleken hos bolagen samt en investeringsstrategi baserad på låga P/BV-multiplar. Urvalen skedde utifrån Spotlight Stock Market, Large Cap och First North. Resultaten som genererades analyserades utifrån avkastning och riskjusterad avkastning samt statistiska tester i from av t-tester genomfördes. Slutsats: Utifrån resultaten kunde en småbolagseffekt inte påvisas. Portföljerna enbart baserade påbolagsstorlek för Spotlight Stock Market och First North presterade avkastningar som alla var lägre änOMXSPI, dock kunde detta inte säkerställas statistiskt. När investeringsstrategin utifrån låga P/BVmultiplar användes kunde portföljen utifrån Spotlight Stock Market påvisa en högre avkastning än OMXSPI, dock kunde denna inte heller statistiskt säkerställas. Gällande den riskjusterade avkastningen kunde enbart en av portföljerna baserade på Spotlight Stock Market och First Northpåvisa en riskjusterad överavkastning. Portföljerna baserade på Large Cap påvisade både högre avkastning än småbolagsportföljerna samt riskjusterad överavkastning. / Background: One of the most common aspects when investing is to achieve high returns while maintaining the risk at low levels. Risks associated with firms vary due to different factors of which oneis the size. In order to reach satisfactory returns many investment strategies have been presented with the goal of outperforming the market and thus contrasts the efficient market hypothesis. Through investments in firms with low market value of equity, and in combination with other investment strategies, it is of interest to investigate whether outperforming the market is possible. Purpose: The purpose of this study is to analyze whether a Small firm effect on the Spotlight Stock Market and First North during the years 2007-2019 could be proven and if it is possible in combination with investment strategies to generate risk-adjusted returns. Methodology: In order to complete the study a quantitative design in combination with a deductive design was used. Eight different portfolios were compiled based on the size of the companies as well as an investment strategy based on low P/BV-multiples. The samples were chosen from Spotlight StockMarket, Large Cap and First North. The results that were generated were analyzed based on return, risk-adjusted return and statistical test in the form of t-tests. Conclusion: The results from this study show that a Small Firm effect could not be established. The portfolios solely based on company size for Spotlight Stock Market and First North all performed a lower return than OMXSPI, though this could not be statistically confirmed. When an investment strategy based on low P/BV-multiples was used the portfolio based on Spotlight Stock Market showed a return higher than OMXSPI, though this could not be statistically confirmed either. Regarding the risk-adjusted excess return only one of the portfolios based on Spotlight Stock Market and First North could prove a risk-adjusted excess return. The portfolios based on Large Cap both proved a higher return than the small firm portfolios as well as risk-adjusted excess return.

This dissertation collects three essays which deal with financial development in emerging markets. Owing to the appliance of different econometric methods on several data sets, insights in the behavior of and the impacts from financial markets are generated. Usually, the financial markets in emerging countries are characterized by the presence of credit constraints. In the first chapter it is shown that the financial development in 19th century Germany generally affected the economy in a positive way. Additionally, when different economic sectors are under investigation, it is revealed that the reaction due to financial development is not homogeneously across the sectors. A structural vector autoregression (VAR) framework is applied to a new annual data set from 1870 to 1912 that was initially compiled by Walther Hoffmann (1965). With respect to the literature, the most important difference of this analysis is the focus on different sectors in the economy and the interpretation of the results in the context of a two-sector growth model. It is revealed that all sectors were affected significantly by shocks from the banking system. Interestingly, this link is the strongest in sectors with small or non-tradable-goods-producing firms, such as construction, services, transportation and agriculture. In this regard, the growth patterns in 19th century Germany are reminiscent to those in today's emerging markets. The second chapter deals with the integration of the stock markets of mainland China with those of the United States and Hong Kong. Market integration and the resulting welfare gains as risk sharing, increasing investment and growth benefits has become a central topic in international finance research. This chapter investigates stock market integration after stock market liberalization which is assessed by spillover effects from Hong Kong and the United States to Chinese stock market indices. Dividing the sample in pre- and post-liberalization phases, causality in variance procedure is applied using four mainland China stock market indices, two indices of the stock exchange in Hong Kong and the Dow Jones Industrials index in the main part. Evidence of global and regional integration is found, but no evidence for increasing integration after the partial opening of the Chinese stock markets, neither with Hong Kong nor with the United States. Based on the idea presented in the first chapter, the third chapter examines one of today's emerging markets. As China is experiencing remarkable economic growth in the recent decades, it is analyzed if and to what extent the ongoing deregulations in the financial system contribute to this development. Structural VARs for gross domestic product as well as for sectoral output data in conjunction with two different bank lending variables are applied. It is indicated that China is positive affected by financial development and that all sectors benefit from domestic bank lending enlargements but to different degrees. Especially in the sectors where mainly state-owned enterprises are represented - such as construction, trade and transportation - shocks in bank lending have a strong positive influence while sectors where private enterprises are prevalent, seem to be more credit constrained.

Bakgrund Ett ökat engagemang har under de senaste åren noterats på den svenska börsen och intresset för aktier är det starkaste sedan år 2004. Forskning har tidigare studerat engagemanget på börsen utifrån demografiska faktorer, däremot identifieras en kunskapslucka som medföljt med det ökade engagemanget där en ny och utvecklad empirisk grupp noterats. Syfte För att bidra till en förståelse om engagemanget på den svenska börsen syftar denna studie till, med hjälp av en kvantitativ metod, att kartlägga engagemanget på den svenska börsen utifrån de demografiska faktorerna kön, ålder och utbildningsnivå. Metod Denna studie har tillämpat en kvantitativ metod med en deduktiv ansats och en tvärsnittsdesign. Det empiriska underlaget har samlats in med hjälp av en digital enkätundersökning där ett sannolikhetsurval applicerats. Materialet bestod av användbara svar från 116 respondenter som vidare har analyserats med hjälp av korrelationsmatriser, bi- och multivariata regressionsanalyser. Resultat Analysen som genomförts påvisade att kön har ett negativt signifikant samband med engagemang på den svenska börsen, medan ålder och utbildningsnivå inte har visat på ett signifikant samband till variabeln. Slutsats Ett negativ signifikant samband har identifierats där män konstateras ha ett högre engagemang på den svenska börsen än kvinnor. Därutöver har inget ytterligare samband fastställts, ålder och utbildningsnivå kan alltså inte användas vid kartläggning av engagemang på den svenska börsen. / Background In recent years, an increased level of enthusiasm on the Swedish stock market has been recognized, which has not been noted since 2004. While prior research has explored the engagement of individuals on the stock market based on demographic factors, a knowledge gap has been identified with regards to this increased engagement. Specifically, a new and evolved empirical group has been noticed, which warrants further exploration. Purpose In order to contribute to an analysis of the engagement on the Swedish stock market, this study aims to, with the use of a quantitative method, map the engagement on the Swedish stock market through the demographic factors gender, age and education level. Methodology This study utilizes a quantitative method with a deductive approach and a cross-sectional design. The empirical data has been collected through a digital survey where a probability sample was used. The material consisted of 116 appropriate answers, which were further analyzed with the assistance of correlation matrices, bi- and multivariate regression analyses. Results The analysis conducted demonstrated a negative significance between gender and engagement on the Swedish stock market, while no significance could be observed between the variable studied and, age and education level. Conclusion The negative significance that has been identified indicates that men have a higher engagement rate on the Swedish stock market compared to women. No further relation has been established, resulting in that age and level of education cannot be used to map the engagement on the Swedish stock market.

The research undertaken aims to contribute to the debate about market efficiency and market volatility in an Islamic context. The research relates to the Amman Stock Exchange (ASE) and covers the period 1992 to 2007. It undertakes quantitative analysis involving two key elements: first, testing for random walk and calendar anomaly effects in market returns and, second, modelling volatility in market returns. The thesis applies a series of standard econometric and statistical techniques to this issue. The key ‘novel’ contributions of this study relate to the focus on Islamic religious holiday effects and also the application of behavioural finance theoretical models to explain the findings in terms of the influence of social mood (mood misattribution) effects. These are approaches that have not been previously applied in the literature within an Islamic context. The author argues that the econometric and statistical techniques applied are ‘fit for purpose’. Standard methods are applied; however, these are applied in ‘novel’ ways in parts of the thesis. For example, moving-date calendar effects are modelled for the first time and the modelling of volatility makes use of interaction effects to explore the impact of interactions between different mood-influencing variables. The study begins by identifying that the ASE index returns do not follow a Random Walk. It then goes on to identify day-of-the-week effects. First trading day of the week effects found in relation to the first trading day that follows the Muslim holy day of Friday. Monthly calendar effects were also found. January or turn-of-the-year effects were found in the ASE similar to those found previously in some Western markets. However, the largest monthly effects were found in relation to the holy month of Ramadan. Most significantly, Ramadan was found to be the only month where the average daily returns were both statistically different from the other months in the year and also positive. This, it is argued in the thesis, is due to social mood (or mood misattribution) effects. The research looks beyond informational efficiency and develops a number of ‘novel’ contributions to research in this area in terms of both the empirical findings and the behavioural finance-related interpretation of these findings, as well as the influence of Islamic ethics in Amman’s stock market returns. The thesis also examines the relationship between seven behavioural mood-proxy variables and stock market returns. Fama (1991) argues that efficiency and volatility are unrelated. In this thesis, however, evidence is uncovered which suggests that this may not be the case. High levels of volatility were found at the start and at the end of the Ramadan holy festival; this volatility, it is argued, is related to social mood. This issue is examined further by exploring previously unstudied interactions between mood-related Ramadan effects and mood-related weather and biorhythmic effects. The results of this thesis, the author believes, provide strong evidence for the existence of Muslim religion investment decision biases associated with social mood effects (mood misattribution). It is argued that these social mood effects in the case of Jordan relate mainly to Islamic ethics and cultural issues, as they are found predominantly during the Ramadan religious holiday. Despite the existence of decision biases within the ASE, no profitable trading anomaly opportunities were identified. This may be due, in part, to Jordan having high trading transaction costs. It is possible, however, that profitable trading opportunities related to Islamic holidays may exist in countries that follow stricter religious observance. The author believes that there is an opportunity to extend this research to countries such as Bahrain.

本論文共一冊，約五萬字，分為四，如下： 一、導論：探討以台灣證券市場為一單元的國際證券組合對國內外投資個人，與機構 有何特別涵意。 二、文獻探討：以兩個理論觀點去探討之，其一視國際資本市場為一區隔市場，另一 則認為此一市場為整合市場，且整合趨勢愈來愈明顯。 三、研究方法：（一）從整合模式上去探討台灣證券市場的報酬率與世界資本市場報 酬率的相關性，以試圖驗證「整合論」。 （二） a•從區隔觀點，去分析台灣股價報酬率與世界其它十四個國家的關係。 b•進一步分析是否可用產業而增加此一關係變數的解釋能力。 四、研究結果：根據預測： （一）整合模式之驗證將不明顯。 （二）區隔測試應該顯示出我國股市與美國、日本的相關性為最大。 香港以應名列前矛。這主要是依據我國經貿關係之密切程度而作的推論。

無 / This paper investigates the long term performance of Taiwan stock market from 1967 to 2008. We obtain the total return of Taiwan Stock Exchange Capitalization Weighted Stock Index (“TAIEX”) by adjusting cash dividends. Prior to Dec 31st, 2008, the adjusted TAIEX (AdTAIEX) become 16,088.49 is 3.5 times than 4591.23, the TAIEX it reach. Based on annualized rate of return, AdTAIEX has 13.069% greater than 9.743% of TAIEX. Investors not only care about the rate of return of their portfolio, but the real purchasing power they have. Based on correction of inflation, AdTAIEX only has 8.25% annually, with 4.819% erosion by inflation. TAIEX only has 5.07% annually, with 4.673% erosion by inflation. Also, we create a timing strategy according to the varying of past variance that may help “buy and hold”strategy.