A complex system is one with many parts, whose behaviors are strongly dependent on each other. There are two interesting questions about complex systems. One is to understand how to recover the true structure of a complex system from noisy data. The other is to understand how the system interacts with its environment. In this thesis, we address these two questions by studying two distinct complex systems: dynamic systems and market microstructure. To address the first question, we focus on some nonlinear dynamic systems. We develop a novel Bayesian statistical method, Gaussian Emulator, to estimate the parameters of dynamic systems from noisy data, when the data are either fully or partially observed. Our method shows that estimation accuracy is substantially improved and computation is faster, compared to the numerical solvers. To address the second question, we focus on the market microstructure of hidden liquidity. We propose some statistical models to explain the hidden liquidity under different market conditions. Our statistical results suggest that hidden liquidity can be reliably predicted given the visible state of the market. / Statistics

Diese Arbeit enthält vier Aufsätze über die Beziehungen zwischen Handelsstrategien auf Aktienmärkten und dem Zustand des Marktes. Es werden ökonometrische Methoden angewendet um den Markteinfluss von limit order Platzierungen, die Eigenschaften von limit ordern sowie die Verwendung von versteckten ordern zu analysieren. Im Kapitel 1 quantifizieren wir die Effekte der limit order Platzierung auf Preisquotierungen am Börsenplatzes Euronext. Wir zeigen, dass eine limit order signifikante Informationen enthält und illustrieren inwieweit ihr Einfluss auf den Markt von ihren Charakteristika und dem Zustand des Orderbuchs abhängt. Das Kapitel 2 enthält empirische Resultate über die limit order Aktivität und den Markteinfluss von limit orders an der New Yorker NASDAQ Börse. Wir dokumentieren, dass Marktteilnehmer die Platzierung von limit orders mit kleinen Volumina präferieren, diese aber sofort nach ihrem Einsatz wieder löschen. Basierend auf der geschätzten Marktauswirkung einer limit order schlagen wir eine Methode zur Prognose ihres optimalen Volumens vor. Im Kapitel 3 werden die limit order-Strategien von Marktteilnehmern in intransparenten Märkten untersucht. Wir zeigen, dass die Position der sogenannten versteckten Liquidität im Orderbuch von diversen Variablen abhängt, die den Zustand des Marktes beschreiben. Die Daten suggerieren, dass Händler die Platzierung sogenannter hidden orders im Hinblick auf günstige Liquidität am Markt und dem "picking-off"-Risiko ausbalancieren. Im letzten Kapitel 4 präsentieren wir ein Softwaresystem zur Rekonstruktion von Orderbüchern und zur Extrahierung von Orderflussinformationen aus message stream Daten. Die Basismodule des Systems basieren auf allgemeinen Orderbuch-Ereignissen. Sie sind abstrakt gehalten und können so einfach auf beliebige Märkte mit elektronischen Orderbüchern angewendet werden. / In four essays, this thesis examines the interaction between traders'' strategies and the state of market by the econometric analysis of maket impact of limit order submission, the typical properties of order flow and the traders'' usage of hidden orders. Chapter 1 quantifies short-term and long-term effects of limit order submissions on quotes in Euronext. We show that limit orders have significant information content and the maginitude of their impact on the quotes depends on both the order''s characteritics and the state of limit order books (LOBs). Chapter 2 provides new empirical evidence on order submission activities and market impacts of limit orders at NASDAQ. We find that traders dominantly submit small size limit orders and cancell most of them immediately after submission. Based on the estimated market impact of orders, we propose a method to predict the optimal size of a limit order conditional on its position in the LOB and the desired impact. Chapter 3 analyzes traders'' decisions on using undisclosed orders in opaque markets. Our empirical findings show that market conditions affect traders'' order submission strategies and suggest that traders balance their hidden order placements to compete for the provision of liquidity and protect themselves against picking-off risk. Chapter 4 presents a program framework for reconstructing LOBs as well as extracting order flow information from message stream data. We design the basic modules of the system in an abstract layer based on common order events in limit order markets, so that it can be easily adapted to data at any limit order markets.

An den Handelsbörsen der Welt, hat der Anteil unsichtbarer Luidität in den letzten Jahren dramatisch zugenommen. Obwohl dieser Trend zunehmend in den Fokus regulatorischer Debatten und akademischer Dikussionen rückt, sind sich Forscher und die Aufsichtsbehörden über die Implikationen und entsprechende regulatorische Maßnahmen uneins. In der vorliegenden Arbeit, werden die damit verbundenen Fragestellungen in drei separaten Kapiteln theoretisch und empirisch untersucht. Mit Hilfe eines speziellen NASDAQ Datensatzes, werden in Kapitel 1 die Marktfaktoren, die unsichtbaren Liquidität begünstigen sowie den Einfluß, den unsichtbare Liquidät auf Märkte ausübt, empirisch ausgewertet. Wir zeigen, daß die Querschnittsvariation unsichtbarer Liquidität entlang des Aktienuniversums in einem hohen Maße durch sichtbare Markteigenschaften erklärt wird. Wir zeigen, daß unsichtbare Order gegenüber sichtbaren Ordern signifikant stärkere Preisfluktuationen hervorrufen. Unsere Resultate geben Grund zu der Annahme, daß Märkte mit hoher unsichtbarer Liquidät volatiler sind und höheren Marktreibungen ausgesetzt sind. In Kapitel 2 entwickeln wir ein strukturelles Handelsmodell und untersuchen die optimale Handelsstrategie mit unsichtbaren Ordern. In diesem Rahmen leiten wir für verschiedene Marktspezifikationen explizite Charakterisierungen der sogenannten optimalen Exposure-Größe her. Unter anderem zeigen wir, daß der Einsatz unsichtbarer Order Transaktionskosten signifikant reduzieren kann. In Kapitel 3 entwickeln wir ein dynamisches, Gleichgewichtsmodell in einem Limitorderbuchmarkt. Innerhalb dieses theoretischen Rahmens können die empirischen Beobachtungen des ersten un zweiten Kapitels rationalisiert werden. Insbesondere zeigen wir daß große versteckte Order Marktineffizienzen hervorrufen und Preisfluktuationen verstärken, indem sie die Koordination zwischen Angebots- und Nachfrageseite schwächen können. / In recent years, the proliferation of hidden liquidity in financial markets has increased dramatically and shifted to the center regulatory debates and market micro-structure panels. Yet investors, scientists and policy makers are at odds about its implications and the adequate regulatory responses. This thesis addresses these issues in three separate chapters on both empirical and theoretical grounds. Chapter 1 provides an empirical investigation of the determinants and impact of hidden order submissions. We report that the cross-sectional variation of hidden liquidity is well explained by observable market characteristics. Second, our results suggest that the hidden orders generate substantial price reactions. Our results suggests that hidden liquidity increases market volatility and trading frictions. Chapter 2 proposes a structural trading model. We investigate trader’s optimal trading strategies with respect to order-exposure in limit order book markets. The optimal exposure size marks a trade-off between costs and benefits of exposure. Our model provides explicit characterizations of the optimal exposure size for various market specifications. Model parameters and exposure strategies are estimated through high-frequency order book data. Our results suggest that hidden orders can substantially enhance trade performance. Chapter 3 develops a dynamic equilibrium model with a public primary market and an off-exchange trading mechanism. Our theory correctly predicts the key findings of chapter one and two. For instance, we show that large hidden orders cause excess returns and increase market volatility and correctly predict the role of the observable market characteristics in the origination of hidden liquidity.

Neste trabalho, apresentamos um conjunto de fatos empíricos e propriedades estatística de negociações em alta frequência, e discutimos algumas questões gerais comuns a dados de alta frequência tais: como discretização, espaçamento temporal irregular, durações correlacionadas, periodicidade diária, correlações temporais e as propriedades estatísticas dos fluxos de ordens. Logo apresentamos dois modelos da literatura,estilizados para a dinâmica do limit order book. No primeiro modelo os fluxo de ordens é descrito por processos de Poisson independentes, propomos para ele uma forma alternativa da prova de ergodicidade basejada em funções de Lyapunov. O segundo modelo é um modelo reduzido que toma em consideração dinâmicas tipo difusão para os tamanhos do bid e ask, e se foca só nas ordens como melhores preços, e modela explicitamente as cotações do bid e ask na presença de liquidez oculta. E por ultimo, propomos um modelo alternativo para a dinâmica do preço e do spread no limit order book, estudamos o comportamento assintótico do modelo e estabelecemos condições de ergodicidade e transitoridade. Além disso, consideramos a uma família de cadeias de Markov definidos nas sequências de caracteres (strings, ou palavras) com infinito alfabeto e para alguns exemplos inspirados nos modelos de negociações em alta frequência, obtemos condições para ergodicidade, transitoriedade e recorrência nula, para a qual usamos as técnicas de construção de funções Lyapunov. / In this work, we present a set of empirical facts and statistical properties of negotiations at high frequency and discuss some general issues common to high-frequency data such: as discretization, irregular spacing, correlated durations, daily periodicity, temporal correlations and the statistical properties of flows orders. Soon we present two models stylized in the literature for the dynamic limit order book. In the first model the order flow described by separate Poisson processes and we propose it to an alternative form of test ergodicity based on Lyapunov function. The second model is a reduced model that takes into consideration diffusion-type dynamics for the sizes of the bid and ask, and focus only on orders as best price and model explicitly quotes the bid and ask in the presence of hidden liquidity. And finally, we propose an alternative model for the price dynamics and spread in the limit order book, we study the asymptotic behavior of the model and established conditions of ergodicity. Furthermore, we consider the a family of Markov chains defined on the sequences of characters (strings, or words) with infinite alphabet. For some examples inspired by the models of high frequency trading we obtain a conditions for ergodicity, transience and null-recurrence. In order to prove this we use the construction of Lyapunov functions techniques.

Neste trabalho, apresentamos um conjunto de fatos empíricos e propriedades estatística de negociações em alta frequência, e discutimos algumas questões gerais comuns a dados de alta frequência tais: como discretização, espaçamento temporal irregular, durações correlacionadas, periodicidade diária, correlações temporais e as propriedades estatísticas dos fluxos de ordens. Logo apresentamos dois modelos da literatura,estilizados para a dinâmica do limit order book. No primeiro modelo os fluxo de ordens é descrito por processos de Poisson independentes, propomos para ele uma forma alternativa da prova de ergodicidade basejada em funções de Lyapunov. O segundo modelo é um modelo reduzido que toma em consideração dinâmicas tipo difusão para os tamanhos do bid e ask, e se foca só nas ordens como melhores preços, e modela explicitamente as cotações do bid e ask na presença de liquidez oculta. E por ultimo, propomos um modelo alternativo para a dinâmica do preço e do spread no limit order book, estudamos o comportamento assintótico do modelo e estabelecemos condições de ergodicidade e transitoridade. Além disso, consideramos a uma família de cadeias de Markov definidos nas sequências de caracteres (strings, ou palavras) com infinito alfabeto e para alguns exemplos inspirados nos modelos de negociações em alta frequência, obtemos condições para ergodicidade, transitoriedade e recorrência nula, para a qual usamos as técnicas de construção de funções Lyapunov. / In this work, we present a set of empirical facts and statistical properties of negotiations at high frequency and discuss some general issues common to high-frequency data such: as discretization, irregular spacing, correlated durations, daily periodicity, temporal correlations and the statistical properties of flows orders. Soon we present two models stylized in the literature for the dynamic limit order book. In the first model the order flow described by separate Poisson processes and we propose it to an alternative form of test ergodicity based on Lyapunov function. The second model is a reduced model that takes into consideration diffusion-type dynamics for the sizes of the bid and ask, and focus only on orders as best price and model explicitly quotes the bid and ask in the presence of hidden liquidity. And finally, we propose an alternative model for the price dynamics and spread in the limit order book, we study the asymptotic behavior of the model and established conditions of ergodicity. Furthermore, we consider the a family of Markov chains defined on the sequences of characters (strings, or words) with infinite alphabet. For some examples inspired by the models of high frequency trading we obtain a conditions for ergodicity, transience and null-recurrence. In order to prove this we use the construction of Lyapunov functions techniques.