Operações de alta frequência ganharam destaque nos últimos anos, tanto no mercado nacional quanto internacional, e têm atraído a atenção de reguladores, pesquisadores e da mídia. Assim, surgiu a necessidade de estudar o mercado de capitais brasileiro no contexto dos dados em alta frequência. Este estudo preocupa-se em analisar os efeitos dos avanços tecnológicos e novas formas de negociação na qualidade do mercado. Tais pontos são caracterizados pelo HFT. Gomber e Haferkorn (2013) explicam que HFT é um subgrupo das negociações com algoritmos. Os investidores HFTs são caracterizados por negociarem com seu próprio capital, manterem posições por espaços curtos de tempo, pelo alto volume de negociação e por atualizarem as ordens com frequência. A revisão da literatura permitiu delinear o termo e identificar as estratégias adotadas, os impactos positivos e negativos na qualidade de mercado, os riscos advindos da prática e medidas adotadas ou propostas para mitigar esses riscos. A contribuição decorrente das negociações em alta frequência foi analisada empiricamente com ênfase na questão da eficiência informacional do mercado nacional. Para isso, foram utilizados dados intradiários do índice Bovespa, com frequências de observação a partir de 1 minuto. Aplicações do teste de sequência para aleatoriedade e teste de razão de variância de Lo e Mackinlay (1988) evidenciaram um aumento na eficiência do mercado ao longo do período analisado, entre 2007 e 2015, para a frequência de observações de 1 minuto. Foi encontrada relação entre esse ganho em eficiência e o aumento da participação do HFT no mercado. Também foi constatado que o mercado se mostra menos eficiente quando a frequência de observação aumenta e que os ganhos em eficiência são mais acentuados para frequências maiores. Os últimos resultados fortalecem a percepção de que a melhora na eficiência está relacionada diretamente à atuação dos HFTs no mercado, haja vista a característica destes de explorarem ineficiências de preço em frações de segundos. Descreveu-se assim o mercado de capitais nessa era de alta frequência e os impactos do HFT na eficiência de mercado. Tais pontos podem ser colocados como contribuições práticas deste estudo. / High-frequency trading has gained notoriety in recent years and attracted incresing attention among policymakers, researchers and media. This brought about the need for research of high frequency data on brazilian capital market. This study aims to investigate the effects of technological advancements and new forms of trading, specially HFT, on market quality. Gomber and Haferkorn (2013, p. 97) define HFT as a subset of algorithmic trading \"characterized by short holding periods of trading positions, high trading volume, frequent order updates and proprietary trading\". The literature review made it possible to define the term and identify strategies, positive and negative impacts on market quality, risks and ways to mitigate these risks. The contribution arising from HFT was analyzed empirically with an emphasis on price efficiency in the domestic market, using intraday Bovespa index data in different frequencies. Run tests and Lo and Mackinlay (1988) variance ratio tests showed increasing efficiency over the period, between 2007 and 2015, for observations in 1 minute frequency. Relationship between this gain in price efficieny and the growth of HFT market share was found. It was found that the market is less eficiente when higher frequencies are analyzed, and that the efficiency gains are more pronounced for higher frequencies. The last results strengthen the perception that the efficiency gains are directly related to high-frequency trading, given its characteristc of exploring price inefficiencies that last fractions of seconds. The capital market in this high frequency era and the impacts of HFT on market efficiency were described in this study

Submitted by Silvana Teresinha Dornelles Studzinski (sstudzinski) on 2015-10-21T11:05:40Z No. of bitstreams: 1 PAULO A. SCHMIDT_.pdf: 1386765 bytes, checksum: f2caadfc119f6eda2a41f0afb9efe1f1 (MD5) / Made available in DSpace on 2015-10-21T11:05:40Z (GMT). No. of bitstreams: 1 PAULO A. SCHMIDT_.pdf: 1386765 bytes, checksum: f2caadfc119f6eda2a41f0afb9efe1f1 (MD5) Previous issue date: 2015-07-27 / Nenhuma / A previsão de preços de ações é um assunto de grande interesse tanto por parte de agentes de mercado quanto da comunidade científica e acadêmica. Ao mesmo tempo, o problema é considerado como um dos mais desafiadores no tratamento de séries temporais, dada sua natureza altamente dinâmica. Uma ampla gama de estudos propõe-se a abordar o tema. Alguns com resultados bastante promissores fazem uso de Redes Neurais Artificiais (RNAs) do tipo Focused Time Lagged FeedForward Network (FTLFN), as quais apresentam mecanismos de memória capazes de detectar padrões temporais. Em muitos casos, no entanto, as capacidades da rede neural não são devidamente exploradas, limitando-se a testes com um conjunto mínimo de parâmetros. Além disso, a maioria dos estudos de previsões de preços de ações possui como foco períodos de baixa frequência, como dias ou meses. Contudo, devido à facilidade de acesso à informação nas últimas décadas e à automatização das negociações em bolsas de valores, estas são realizadas cada vez mais sob horizontes de curto prazo, como horas, minutos ou segundos. Existe, portanto, a necessidade de se expandir o conhecimento em relação a previsões dentro deste cenário. Neste sentido, este trabalho tem como objetivo uma investigação das reais potencialidades de previsão das FTLFNs sobre preços de ações no período intradiário. Sua memória de curto prazo e tamanho de camada oculta são explorados de forma ampla e aprofundada, através dos quais se buscou identificar o impacto das diferentes configurações nos resultados de acurácia dentro do contexto considerado. Na tentativa de oferecer suporte a melhores previsões, analisa-se também a influência de indicadores da Análise Técnica sobre o modelo. De forma mais geral, procura-se ampliar o entendimento a respeito tanto das capacidades de previsão das redes do tipo FTLFN como de sua empregabilidade em séries temporais financeiras intradiárias, ainda pouco exploradas na literatura. Os resultados obtidos mostram que, assim como investidores humanos, também as FTLFNs são capazes de se beneficiar enormemente de padrões formados pelos históricos dos sinais de entrada, a fim de prover previsões de maior qualidade dentro do contexto proposto neste trabalho. O mesmo não pode ser afirmado a respeito dos indicadores da Análise Técnica escolhidos, uma vez que em sua grande maioria aumentam os erros de previsão. As evidências apresentadas baseiam-se em experimentações sobre diferentes conjuntos de sinais, oferecendo robustez às conclusões alcançadas e permitindo que a metodologia e os resultados sirvam como base para futuras pesquisas relacionadas a previsões dentro de cenários de alta frequência. / Stock price prediction is a subject of great interest for both market agents and scientific and academic community. At the same time, this problem is considered to be one of the most challenging in time series forecasting, due to its highly dynamic nature. A large amount of researches have proposed to address the issue. Some of them, with very promising results, adopt the Focused Time Lagged FeedForward Network (FTLFN), a type of Artificial Neural Network (ANN) that offers memory mechanisms capable of detecting temporal patterns. In many cases, however, the neural network’s capacities are not properly explored, being limited to tests with a minimum set of parameters. Besides, most of the studies on stock price prediction focus on low-frequency periods, such as days or months. On the other hand, due to the ease of access to information in the last decades and the automation of trades in stock market, these are getting more oftenly executed over short-term horizons, like hours, minutes or seconds. Therefore, there is a need to expand the knowledge related to forecasts in this scenario. With that in mind, this research has the objective of investigating the FTLFN’s potential on stock price forecasting over the intraday period. Its short-term memory and hidden layer size are widely and de eply explored, so the impact of different configurations on the accuracy results could be measured. Also, Technical Analysis indicators are built and utilized as input signals to the network, with their possible contributions to stock prediction being verified. From a general perspective, the work proposes the extention of the understanding regarding the FTLFN’s forecasting capabilities, as well as its use with intraday financial time series, which still require further exploration in literature. The obtained results show that, as human investors do, also FLTFNs are capable of taking enormous advantage from input signals’ history on providing better prediction quality within the proposed context. The same cannot be said for the supporting Technical Analysis indicators chosen, since they mostly increase forecasting errors. Evidences are presented based on the experimentation over several sets, bringing robustness to the conclusions and allowing the methodology and the results to serve as base for future researches related to predictions on high-frequency trading scenarios.

This thesis is composed of three chapters which propose some novel approaches to model and forecast financial time series. The first chapter focuses on high frequency financial returns and proposes a quantile regression approach to model their intraday seasonality and dynamics. The second chapter deals with the problem of forecasting the yield curve including large datasets of macroeconomics information. While the last chapter addresses the issue of modelling the term structure of interest rates. The first chapter investigates the distribution of high frequency financial returns, with special emphasis on the intraday seasonality. Using quantile regression, I show the expansions and shrinks of the probability law through the day for three years of 15 minutes sampled stock returns. Returns are more dispersed and less concentrated around the median at the hours near the opening and closing. I provide intraday value at risk assessments and I show how it adapts to changes of dispersion over the day. The tests performed on the out-of-sample forecasts of the value at risk show that the model is able to provide good risk assessments and to outperform standard Gaussian and Student’s t GARCH models. The second chapter shows that macroeconomic indicators are helpful in forecasting the yield curve. I incorporate a large number of macroeconomic predictors within the Nelson and Siegel (1987) model for the yield curve, which can be cast in a common factor model representation. Rather than including macroeconomic variables as additional factors, I use them to extract the Nelson and Siegel factors. Estimation is performed by EM algorithm and Kalman filter using a data set composed by 17 yields and 118 macro variables. Results show that incorporating large macroeconomic information improves the accuracy of out-of-sample yield forecasts at medium and long horizons. The third chapter statistically tests whether the Nelson and Siegel (1987) yield curve model is arbitrage-free. Theoretically, the Nelson-Siegel model does not ensure the absence of arbitrage opportunities. Still, central banks and public wealth managers rely heavily on it. Using a non-parametric resampling technique and zero-coupon yield curve data from the US market, I find that the no-arbitrage parameters are not statistically different from those obtained from the Nelson and Siegel model, at a 95 percent confidence level. I therefore conclude that the Nelson and Siegel yield curve model is compatible with arbitrage-freeness.

More renewable energy leads to higher energy imbalances in the Swedish electric power system. In the same time, the grid capacity is almost reached in some regions which requires an extension of the current grids or a reduction of the power consumption. Demand response could be a key factor for both stabilizing the energy balances and reducing the grid congestion. The aim with this thesis is to analyze the potential incomes that demand response from heat pumps can generate for the balance responsibility parties and the grid operators and evaluate how it would affect the end-consumers.   The investigated local grid that contains of 174 multi-family buildings with heat pumps could reduce its highest peak power with 2,9 MW. This peak power reduction generated a cost reduction of 483 000 SEK per year or 2800 SEK per building per year in reduced penalty fees and power subscription fees. The mFRR market and the power reserve market were determined to be the most suitable markets for using demand response from heat pumps on for the balance responsibility party in the electricity price region SE3. SE3 consists of 10146 multi-family buildings with heat pumps. The mFRR market generated an average income of 2 699 000 SEK per winter season whereas the power reserve market generated a yearly administrative compensation of 1 133 000 SEK per season and 104 000 SEK per call-off. It is important that end-consumers obtain demand-based tariffs or hourly based tariffs to enable a cost reduction from the control system.

Operações de alta frequência ganharam destaque nos últimos anos, tanto no mercado nacional quanto internacional, e têm atraído a atenção de reguladores, pesquisadores e da mídia. Assim, surgiu a necessidade de estudar o mercado de capitais brasileiro no contexto dos dados em alta frequência. Este estudo preocupa-se em analisar os efeitos dos avanços tecnológicos e novas formas de negociação na qualidade do mercado. Tais pontos são caracterizados pelo HFT. Gomber e Haferkorn (2013) explicam que HFT é um subgrupo das negociações com algoritmos. Os investidores HFTs são caracterizados por negociarem com seu próprio capital, manterem posições por espaços curtos de tempo, pelo alto volume de negociação e por atualizarem as ordens com frequência. A revisão da literatura permitiu delinear o termo e identificar as estratégias adotadas, os impactos positivos e negativos na qualidade de mercado, os riscos advindos da prática e medidas adotadas ou propostas para mitigar esses riscos. A contribuição decorrente das negociações em alta frequência foi analisada empiricamente com ênfase na questão da eficiência informacional do mercado nacional. Para isso, foram utilizados dados intradiários do índice Bovespa, com frequências de observação a partir de 1 minuto. Aplicações do teste de sequência para aleatoriedade e teste de razão de variância de Lo e Mackinlay (1988) evidenciaram um aumento na eficiência do mercado ao longo do período analisado, entre 2007 e 2015, para a frequência de observações de 1 minuto. Foi encontrada relação entre esse ganho em eficiência e o aumento da participação do HFT no mercado. Também foi constatado que o mercado se mostra menos eficiente quando a frequência de observação aumenta e que os ganhos em eficiência são mais acentuados para frequências maiores. Os últimos resultados fortalecem a percepção de que a melhora na eficiência está relacionada diretamente à atuação dos HFTs no mercado, haja vista a característica destes de explorarem ineficiências de preço em frações de segundos. Descreveu-se assim o mercado de capitais nessa era de alta frequência e os impactos do HFT na eficiência de mercado. Tais pontos podem ser colocados como contribuições práticas deste estudo. / High-frequency trading has gained notoriety in recent years and attracted incresing attention among policymakers, researchers and media. This brought about the need for research of high frequency data on brazilian capital market. This study aims to investigate the effects of technological advancements and new forms of trading, specially HFT, on market quality. Gomber and Haferkorn (2013, p. 97) define HFT as a subset of algorithmic trading \"characterized by short holding periods of trading positions, high trading volume, frequent order updates and proprietary trading\". The literature review made it possible to define the term and identify strategies, positive and negative impacts on market quality, risks and ways to mitigate these risks. The contribution arising from HFT was analyzed empirically with an emphasis on price efficiency in the domestic market, using intraday Bovespa index data in different frequencies. Run tests and Lo and Mackinlay (1988) variance ratio tests showed increasing efficiency over the period, between 2007 and 2015, for observations in 1 minute frequency. Relationship between this gain in price efficieny and the growth of HFT market share was found. It was found that the market is less eficiente when higher frequencies are analyzed, and that the efficiency gains are more pronounced for higher frequencies. The last results strengthen the perception that the efficiency gains are directly related to high-frequency trading, given its characteristc of exploring price inefficiencies that last fractions of seconds. The capital market in this high frequency era and the impacts of HFT on market efficiency were described in this study

Práca si kladie za cieľ vytvorenie obchodného systému pre intradenné obchodovanie US akciových indexov. Autor sa v teoretických východiskách zameriava na vysvetlenie základných pojmov obchodovania US indexov na intradennej báze a obchodovania všeobecne. V ďalšej kapitole je popísaná súčastná situácia – tvorba obchodného plánu a vysvetlené principy na ktorých je plán založený. V návrhovej časti je predstavený kompletný systém pre intradenné obchodovanie US akciových indexov.

This work deals with the usage of neural network for the purpose of stock market prediction. A basic stock market theory and trading approaches are mentioned at the beginning of this work. Then neural networks and their application are discussed with their deeper description. Similar approaches are referred and finally two new prediction systems are designed. These systems are utilized by proposed trading model and tested on selected data. The results are compared to human and random trading models and new development steps are devised at the end of this work.

L’énergie éolienne, l’une des énergies renouvelables les plus compétitives, est considérée comme une solution qui remédie aux inconvénients de l’énergie fossile. Pour une meilleure gestion et exploitation de cette énergie, des prévisions de sa production s’avèrent nécessaires. Les méthodes de prévisions utilisées dans la littérature permettent uniquement une prévision de la moyenne annuelle de cette production. Certains travaux récents proposent l’utilisation du Théorème Central Limite (TCL), sous des hypothèses non classiques, pour l’estimation de la production annuelle moyenne de l’énergie éolienne ainsi que sa variance pour une seule turbine. Nous proposons dans cette thèse une extension de ces travaux à un parc éolien par relaxation de l’hypothèse de stationnarité la vitesse du vent et la production d’énergie, en supposant que ces dernières sont saisonnières. Sous cette hypothèse la qualité de la prévision annuelle s’améliore considérablement. Nous proposons aussi de prévoir la production d’énergie éolienne au cours des quatre saisons de l’année. L’utilisation du modèle fractal, nous permet de trouver une division ”naturelle” de la série de la vitesse du vent afin d’affiner l’estimation de la production éolienne en détectant les points de ruptures. Dans les deux derniers chapitres, nous donnons des outils statistiques de la détection des points de ruptures et d’estimation des modèles fractals. / The wind energy, one of the most competitive renewable energies, is considered as a solution which remedies the inconveniences of the fossil energy. For a better management and an exploitation of this energy, forecasts of its production turn out to be necessary. The methods of forecasts used in the literature allow only a forecast of the annual mean of this production. Certain recent works propose the use of the Central Limit Theorem (CLT), under not classic hypotheses, for the estimation of the mean annual production of the wind energy as well as its variance for a single turbine. We propose in this thesis, an extension of these works in a wind farm by relaxation of the hypothesis of stationarity the wind speed and the power production, supposing that the latter are seasonal. Under this hypothesis the quality of the annual forecast improves considerably. We also suggest planning the wind power production during four seasons of the year. The use of the fractal model, allows us to find a "natural" division of the series of the wind speed to refine the estimation of the wind production by detecting abrupt change points. Statistical tools of the change points detection and the estimation of fractal models are presented in the last two chapters.

This thesis is among the first market microstructure studies of an index futures market with designated market makers in the academic literature. The purpose of this thesis is to investigate intraday patterns of key variables, the relative size of the components of the quoted bid-ask spread, and the order decisions of uninformed traders, in a continuous dealer market for index futures with market makers. Overall, our findings aim to contribute to a better understanding of the roles of market makers and public customers in price formation. Intraday patterns of financial market variables such as trade price, volume, trade size, quoted spreads, depth, and volatility separately for designated market makers and public customers are examined. The lack of relevant and appropriate data in futures markets, as evidenced by Hasbrouck (2003) and Kurov (2005), has inhibited the growth of market microstructure in futures markets. Individual orders, quotes, trader identification, and transactions from June 2003 to December 2004, for FTSEurofirst 80 and 100 index futures are used in the study. Inclusion of the parties to order execution distinguishes this data set from most other futures microstructure sources. As this thesis is the first known academic study of the extant market microstructure of the FTSEurofirst index futures, the institutional aspects of the trading process for the FTSEurofirst index futures are also explored. An alternative method for estimating three cost components as a proportion of the bid-ask spread is developed. A framework is developed for the order decision process of an uninformed trader for the first time in a futures market with market makers. The results of this thesis may have implications for other financial markets and the field of market microstructure.

本研究主要以“市場效率性”與“套利機會”之概念出發，並使用類似於“日內事件研究(Intraday event-study)”的方式，觀察標的股票漲停後，認購權證市場在各種特定時間內的情況。希望探討(1) 標的股票漲停後，是否其相對應之認購權證亦同時漲停或達到最高點？(2) 標的股票漲停這一事件，認購權證市場是否有利可圖、是否具有套利之機會？(3) 最後並以Logistic迴歸模式，檢定影響標的股票漲停後，認購權證亦漲停之主要因素，完成此研究。而其中，在時間落差與套利利潤的這前兩項研究當中，本文將資料加以分類，將認購權證價格資料分為 “標的股票漲停後，當日所有時間內”、“標的股票漲停後，一小時內”、“標的股票漲停後，兩小時內”以及“標的股票漲停後，市場即將收盤”等四類加以研究。研究結果如下： 1.無論是在價內、價外或接近價平的認購權證當中，標的股票漲停時間與 認購權證最高價(包括漲停價)之時間差皆相當顯著，顯示標的股票市場 與認購權證市場之間價格資訊之傳遞，具有一定顯著性的落差，效率市 場假說並不成立、認購權證市場對於漲停之正面消息的反應顯著落後於 標的股票市場平均約25分鐘。 2.標的股票漲停後，以此漲停作為一事件並操作“在標的股票漲停時，賣 出標的股票並同時買進相同金額、相對應之權證，且在權證達最高價時 將權證賣出並買入股票還券”之策略的投資者，平均大約可以獲得0.6% ~1.3%的套利利潤。 3.將標的股票漲停後認購權證亦漲停之樣本挑出做檢定，其研究結果顯 示，標的股票漲停與認購權證漲停之時間具有相當大的顯著落差。但 在套利利潤的檢定方面，所有對“標的股票漲停、認購權證亦漲停” 之證券操作套利策略之結果，其所得利潤卻皆為0。探究其原因，標的 股票漲停到認購權證漲停這一段時間內，持有此類型(會漲停)證券的投 資者，皆不會將證券賣出，第一筆可以買到的權證就是權證的漲停價 格，故使得套利策略無法執行，無法取得套利利潤。 4.Logistic迴歸模式之結果： (1) 槓桿比率越高的權證由於其認購權證價格相對於其標的股票顯得低了 很多，故使這類權證越容易在標的股票漲停後，達到漲停。 (2) 相較於價外的認購權證，處於接近價平的認購權證較易吸引投資人， 故使這類權證較價外之權證容易在標的股票漲停後，達到漲停。 (3) 一項足以使標的股票市場漲停的正面消息釋出後，相較於大型券商， 小型券商所發行的認購權證較容易在標的股票漲停後，達到漲停。 (4) 由於重設型權證具有重設的特性，使這類認購權證相較於一般型的認 購權證，更容易貼近於其履約價、更接近於價平，故這類的權證相較 於一般權證更容易在標的股票漲停後，達到漲停。 由於以上所述四類特性之權證，雖然其“時間落差”的檢定結果是顯 著的，但在標的股票漲停與權證漲停這一段落差時間內，卻因無法購 得權證而獲得套利利潤，故在操作套利策略時，應將這類權證加以摒 除。