Submitted by Uirá Caiado de Castro (ucaiado@yahoo.com.br) on 2017-08-28T20:17:54Z No. of bitstreams: 1 uira_caiado_tradingRL.pdf: 1000833 bytes, checksum: d530c31d30ddfd98e5978aaaf3170959 (MD5) / Approved for entry into archive by Joana Martorini (joana.martorini@fgv.br) on 2017-08-28T21:06:42Z (GMT) No. of bitstreams: 1 uira_caiado_tradingRL.pdf: 1000833 bytes, checksum: d530c31d30ddfd98e5978aaaf3170959 (MD5) / Made available in DSpace on 2017-08-29T12:42:53Z (GMT). No. of bitstreams: 1 uira_caiado_tradingRL.pdf: 1000833 bytes, checksum: d530c31d30ddfd98e5978aaaf3170959 (MD5) Previous issue date: 2017-08-24 / O presente estudo propõe o uso de um modelo de aprendizagem por reforço para derivar uma estratégia de trading em taxa de juros diretamente de dados históricos de alta frequência do livro de ofertas. Nenhuma suposição sobre a dinâmica do mercado é feita, porém é necessário criar um simulador com o qual o agente de aprendizagem possa interagir para adquirir experiência. Diferentes variáveis relacionadas a microestrutura do mercado são testadas para compor o estado do ambiente. Funções baseadas em P&L e/ou na coerência do posicionamento das ofertas do agente são testadas para avaliar as ações tomadas. Os resultados deste trabalho sugerem algum sucesso na utilização das técnicas propostas quando aplicadas à atividade de trading. Porém, conclui-se que a obtenção de estratégias consistentemente lucrativas dependem muito das restrições colocadas na aprendizagem. / The present study proposes the use of a reinforcement learning model to develop an interest rate trading strategy directly from historical high-frequency order book data. No assumption about market dynamics is made, but it requires creating a simulator wherewith the learning agent can interact to gain experience. Different variables related to the microstructure of the market are tested to compose the state of the environment. Functions based on P&L and/or consistency in the order placement by the agent are tested to evaluate the actions taken. The results suggest some success in bringing the proposed techniques to trading. However, it is presumed that the achievement of consistently profitable strategies is highly dependent on the constraints placed on the learning task.

La thèse se concentre sur l’étude des impacts de la directive européenne concernant les Marchés d’Instruments Financiers (MIF) et de sa révision (MIF II et MiFIR) sur le processus de découverte des prix. Selon Schreiber et Schwartz (1986), celui-ci est défini comme l´incorporation de l´information nouvelle dans le prix des actifs et la recherche de l´équilibre par le marché. Cette directive clé a pour objectif d´augmenter la concurrence et l´efficience au niveau des marchés européens tout en assurant la protection des investisseurs, ceci via une augmentation de la transparence et une exigence de politique de meilleure exécution des ordres de la part des firmes d´investissement. Plus particulièrement, l´étude se focalise sur les actions françaises du CAC40 qui peuvent désormais être échangées en dehors du marché national réglementé, Euronext Paris. Les plateformes multilatérales de trading, les internalisateurs systématiques et les dark pools sont des alternatives qui ont été introduites par la directive. En l´absence de consolidation du marché européen dans son ensemble et en présence d’une fragmentation spatiale des ordres de bourse, le risque est que certaines places d´échanges reçoivent plus d´ordres d´achat et d´autres, plus d´ordres de vente. Certains pensent que la technologie devrait lier des marchés spatialement fragmentés. Cependant, si suffisamment de flux d´ordres est retiré du marché réglementé et transparent, ce dernier pourrait ne plus assurer la découverte des prix car les prix et les quantités d´équilibre n´auraient pas été découverts par le marché dans son ensemble. De plus, même en présence d´un marché consolidé au niveau spatial, une fragmentation temporelle peut subsister. Elle correspond à la fracturation du flux d’ordres dans le temps, rendant la rencontre des ordres d´achat et de vente plus compliquée. […] La thèse apporte tout d’abord un éclairage sur les enjeux et les implications de la directive sur l’efficience des marchés européens. Dans le premier chapitre, nous proposons un cadre d´évaluation de la directive. Une sélection des principaux travaux académiques est réalisée dans le domaine de la microstructure des marchés afin d´identifier les problématiques restant sans réponse et les enjeux pour sa révision en cours (MIF II). Ensuite, une revue de la littérature sur le processus de découverte des prix du marché est opérée par la mise en lumière des principaux travaux théoriques, méthodologiques et empiriques. Les deux principales fonctions d´un marché sont de fournir de la liquidité et de permettre la découverte des prix. Cependant, la fonction de découverte des prix a souvent été un objectif de régulation négligé par rapport aux objectifs de transparence et de concurrence. […] / The thesis focusses on the impacts of the European Markets in Financial Instruments Directive (MiFID) and its revisions (the MiFID II and the MiFIR) on the price discovery process. According to Schreiber and Schwartz (1986), the price discovery process is defined as the incorporation of new information into the prices of assets and the search for an equilibrium by the market participants. This key directive aims to increase competition and efficiency at the European level without neglecting investor protection by increasing transparency and by requiring a best execution policy for the execution of client orders from investment firms.The study specifically highlights the CAC40 stocks, which, with the implementation of the MiFID, can be exchanged outside the regulated domestic market (Euronext Paris). The directive has introduced Multilateral Trading Facilities (MTFs), Systematic Internalisers (SI) and Dark Pools as alternative trading venues.In absence of an overall consolidation of the European market and in presence of a spatial fragmentation of orders, there is a risk that some exchange places may receive more buy orders and others more sell orders. Technology should bind spatially fragmented markets; however, if enough of the order flow were removed from the regulated and transparent market, it would be unable to ensure the price discovery because the equilibrium prices and quantities would not befound by the overall market. In addition, even in the presence of a consolidated market at the spatial level, temporal fragmentation may still exist. […]The first chapter studies the challenges and the implications of the MiFID on the efficiency of the European financial markets. This research proposes a regulatory framework to assess the directive. A selection of the principal academic work in the microstructure research area has been carried out in order to identify the remaining unanswered issues and challenges for the current revision of the MIF. The second chapter proposes a literature review of the concept of price discovery by highlighting the principal theoretical, methodological and empirical academic research. The two main functions of a market are to provide liquidity and to allow price discovery. However, the price discovery function has often been a neglected regulatory objective in comparison to transparency and competition objectives. It is important to assess the impacts of fragmentation on the quality of the market after the implementation of the MiFID. The object of study is the price discovery accuracy in the post-crisis context of more high-frequency and algorithmic trading. At this level, the thesis first offers a quantification of the degree of spatial and temporal fragmentation of CAC40 shares in the post-MiFID context. This study shows an increase in fragmentation. Furthermore, the quality of the market is evaluated from a price discovery perspective through the study of an indicator developed by Ozenbas et al. (2002, 2011) called the normalised volatility ratio. The study confirms the existence of a price discovery risk at the opening of the market before and after the implementation of MiFID. The potential causes of price discovery accuracy have been studied using three types of variables that characterise each transaction. The number of ransactions and the proportion of high-frequency traders on the buy side for the first half-hour of the day are significant variables for price discovery accuracy. In the post-MiFID scenario, spatial fragmentation does not significantly affect the market quality of CAC40 shares. At this level, temporal fragmentation seems to be a greater determinant. […]

The thesis introduces a framework for analysing market monitoring and surveillance systems in order to provide a common foundation for researchers and practitioners to specify, design, implement, compare and evaluate such systems. The proposed framework serves as a reference map for researchers and practitioners to position their work in the context of market monitoring and surveillance, resulting in a useful instrument for the analysis, testing and management of such systems. More specifically, the thesis examines the new requirements for the operation of financial markets, the role of technologies, the recent consultations on the structure and governance of EU and US markets, as well as, future usage scenarios and emerging technologies. It examines the context in which market monitoring and market surveillance systems are currently been used. It reports on their processes, performance, and on the organisational and regulatory environments in which they exist. Furthermore, it develops a set of taxonomies which cover the majority of the concepts of market manipulation, market monitoring, market surveillance, entities, technologies and actors that are relevant for the work in this thesis. Building on the gaps and limitations of the current systems, it proposes a new framework following the Design Science methodology. The usefulness of the framework is evaluated through four critical case studies, which not only help to understand with practical exercises the way how markets monitoring and surveillance systems work, but also to investigate their weaknesses, potential evolution and ways to improve them. For each case study, the thesis develops a fully working prototype tested using a sample prosecution case and evaluated in terms of the appropriateness and suitability of the proposed framework. Finally, implications relating to policies, procedures and future market structures are discussed followed by suggestions for future research.

Information technology forms an important part of the world and algorithmic trading has already become a common concept among traders. The High Frequency Trading (HFT) requires use of special hardware accelerators which are able to provide input response with sufficiently low latency. This master's thesis is focused on design and implementation of an architecture for order book building, which represents an essential part of HFT solutions targeted on financial exchanges. The goal is to use the FPGA technology to process information about an exchange's state with latency so low that the resulting solution is effectively usable in practice. The resulting architecture combines hardware and software in conjunction with fast lookup algorithms to achieve maximum performance without affecting the function or integrity of the order book.

Cette thèse aborde différents aspects de la modélisation de la microstructure du marché et des problèmes de Market Making, avec un accent particulier du point de vue du praticien. Le carnet d’ordres, au cœur du marché financier, est un système de files d’attente complexe à haute dimension. Nous souhaitons améliorer la connaissance du LOB pour la communauté de la recherche, proposer de nouvelles idées de modélisation et développer des applications pour les Market Makers. Nous remercions en particuler l’équipe Automated Market Making d’avoir fourni la base de données haute-fréquence de très bonne qualité et une grille de calculs puissante, sans laquelle ces recherches n’auraient pas été possible. Le Chapitre 1 présente la motivation de cette recherche et reprend les principaux résultats des différents travaux. Le Chapitre 2 se concentre entièrement sur le LOB et vise à proposer un nouveau modèle qui reproduit mieux certains faits stylisés. A travers cette recherche, non seulement nous confirmons l’influence des flux d’ordres historiques sur l’arrivée de nouveaux, mais un nouveau modèle est également fourni qui réplique beaucoup mieux la dynamique du LOB, notamment la volatilité réalisée en haute et basse fréquence. Dans le Chapitre 3, l’objectif est d’étudier les stratégies de Market Making dans un contexte plus réaliste. Cette recherche contribueà deux aspects : d’une part le nouveau modèle proposé est plus réaliste mais reste simple à appliquer pour la conception de stratégies, d’autre part la stratégie pratique de Market Making est beaucoup améliorée par rapport à une stratégie naive et est prometteuse pour l’application pratique. La prédiction à haute fréquence avec la méthode d’apprentissage profond est étudiée dans le Chapitre 4. De nombreux résultats de la prédiction en 1- étape et en plusieurs étapes ont retrouvé la non-linéarité, stationarité et universalité de la relation entre les indicateurs microstructure et le changement du prix, ainsi que la limitation de cette approche en pratique. / This thesis addresses different aspects around the market microstructure modelling and market making problems, with a special accent from the practitioner’s viewpoint. The limit order book (LOB), at the heart of financial market, is a complex continuous high-dimensional queueing system. We wish to improve the knowledge of LOB for the research community, propose new modelling ideas and develop concrete applications to the interest of Market Makers. We would like to specifically thank the Automated Market Making team for providing a large high frequency database of very high quality as well as a powerful computational grid, without whom these researches would not have been possible. The first chapter introduces the incentive of this research and resumes the main results of the different works. Chapter 2 fully focuses on the LOB and aims to propose a new model that better reproduces some stylized facts. Through this research, not only do we confirm the influence of historical order flows to the arrival of new ones, but a new model is also provided that captures much better the LOB dynamic, notably the realized volatility in high and low frequency. In chapter 3, the objective is to study Market Making strategies in a more realistic context. This research contributes in two aspects : from one hand the newly proposed model is more realistic but still simple enough to be applied for strategy design, on the other hand the practical Market Making strategy is of large improvement compared to the naive one and is promising for practical use. High-frequency prediction with deep learning method is studied in chapter 4. Many results of the 1-step and multi-step prediction have found the non-linearity, stationarity and universality of the relationship between microstructural indicators and price change, as well as the limitation of this approach in practice.

在這份研究中，我們提出一個類分散式神經網路框架，此框架為高頻交易系統研究下之子專案。在系統中，我們透過資料探勘程序發掘財務時間序列中的模式，其中所採用的資料探勘演算法之一即為類神經網路。我們實作一個在分散式平台上訓練類神經網路的框架。我們採用Apache Spark來建立底層的運算叢集，因為它提供高效能的記憶體內運算（in-memory computing）。我們分析一些分散式後向傳導演算法（特別是用來預測財務時間序列的），加以調整，並將其用於我們的框架。我們提供了許多細部的選項，讓使用者在進行類神經網路建模時有很高的彈性。 / In this research, we introduce a distributed framework of artificial neural network (ANN) as a subproject under the research of a high-frequency trading (HFT) system. In the system, ANNs are used in the data mining process for identifying patterns in financial time series. We implement a framework for training ANNs on a distributed computing platform. We adopt Apache Spark to build the base computing cluster because it is capable of high performance in-memory computing. We investigate a number of distributed backpropagation algorithms and techniques, especially ones for time series prediction, and incorporate them into our framework with some modifications. With various options for the details, we provide the user with flexibility in neural network modeling.

Market making – the process of simultaneously and continuously providing buy and sell prices in a financial asset – is rather complicated to optimize. Applying reinforcement learning (RL) to infer optimal market making strategies is a relatively uncharted and novel research area. Most published articles in the field are notably opaque concerning most aspects, including precise methods, parameters, and results. This thesis attempts to explore and shed some light on the techniques, problem formulations, algorithms, and hyperparameters used to construct RL-derived strategies for market making. First, a simple probabilistic model of a limit order book is used to compare analytical and RL-derived strategies. Second, a market making agent is trained on a more complex Markov chain model of a limit order book using tabular Q-learning and deep reinforcement learning with double deep Q-learning. Results and strategies are analyzed, compared, and discussed. Finally, we propose some exciting extensions and directions for future work in this research field. / Likviditetsgarantering (eng. ”market making”) – processen att simultant och kontinuerligt kvotera köp- och säljpriser i en finansiell tillgång – är förhållandevis komplicerat att optimera. Att använda förstärkningsinlärning (eng. ”reinforcement learning”) för att härleda optimala strategier för likviditetsgarantering är ett relativt outrett och nytt forskningsområde. De flesta publicerade artiklarna inom området är anmärkningsvärt återhållsamma gällande detaljer om de tekniker, problemformuleringar, algoritmer och hyperparametrar som används för att framställa förstärkningsinlärningsbaserade strategier. I detta examensarbete så gör vi ett försök på att utforska och bringa klarhet över dessa punkter. Först används en rudimentär probabilistisk modell av en limitorderbok som underlag för att jämföra analytiska och förstärkningsinlärda strategier. Därefter brukas en mer sofistikerad Markovkedjemodell av en limitorderbok för att jämföra tabulära och djupa inlärningsmetoder. Till sist presenteras även spännande utökningar och direktiv för framtida arbeten inom området.

金融市場是個變化莫測的環境，看似隨機，在隨機中卻隱藏著某些特性與關係。不論是自然現象中的氣象預測或是金融領域中對下一時刻價格的預測, 都有相似的複雜性。 時間序列的預測一直都是許多領域中重要的項目之一, 金融時間序列的預測也不例外。在本論文中我們針對金融時間序列的非線性與非穩態關係引入類神經網路(ANNs) 與集合經驗模態分解法(EEMD), 藉由ANNs處理非線性問題的能力與EEMD處理時間序列信號的優點，並進一步與傳統上使用於金融時間序列分析的自回歸滑動平均模型(ARMA)進行複合式的模型建構，引入燭型圖概念嘗試進行高頻下的台指期貨TAIEX交易。在不計交易成本的績效測試下本研究的高頻交易模型有突出的績效，證明以ANNs、EEMD方法與ARMA組成的混合式模型在高頻時間尺度交易下有相當的發展潛力，具有進一步發展的價值。在處理高頻時間尺度下所產生的大型數據方面，引入平行運算架構SPMD(single program, multiple data)以增進其處理大型資料下的運算效率。本研究亦透過分析高頻時間尺度的本質模態函數(IMFs)探討在高頻尺度下影響台指期貨價格的因素。 / Financial market is complex, unstable and non-linear system, it looks like have some principle but the principle usually have exception. The forecasting of time series always an issue in several field include finance. In this thesis we propose several version of hybrid models, they combine Ensemble Empirical Mode Decomposition (EEMD), Back-Propagation Neural Networks(BPNN) and ARMA model, try to improve the forecast performance of financial time series forecast. We also found the physical means or impact factors of IMFs under high-frequency time-scale. For processing the massive data generated by high-frequency time-scale, we pull in the concept of big data processing, adopt parallel computing method ”single program, multiple data (SPMD)” to construct the model improve the computing performance. As the result of backtesting, we prove the enhanced hybrid models we proposed outperform the standard EEMD-BPNN model and obtain a good performance. It shows adopt ANN, EEMD and ARMA in the hybrid model configure for high-frequency trading modeling is effective and it have the potential of development.