In this study we examine the notion of applicability of cointegration statistical arbitrage in Taiwan stock, electronic and financial index future. We form the trading pairs by construction the cointegration relation pairs in the same industry and the same type of business. The basic concept we applied in this way is that market neutral, and contrarian investment. We execute three different kind of pairs. They are individual stock vs. stock pairs, Finance Sector Index Futures and financial stocks, and Electronic and Finance Sector Index Futures vs. Electronic and Financial stock portfolio. The results from the three different kind of combination are all showing the feasibility. of our statistical model.

>Magister Scientiae - MSc / Purpose:This research investigates the existence of long-term equilibrium relationships among the stock market indices of Brazil, Russia, India, China and South Africa (BRICS). It further investigates cointegrated stock pairs for possible implementation of statistical arbitrage trading techniques.Design:We utilize standard multivariate time series analysis procedures to inspect unit roots to assess stationarity of the series. Thereafter, cointegration is tested by the Johansen and Juselius (1990) procedure and the variables are interpreted by a Vector Error Correction Model (VECM). Statistical arbitrage is investigated through the pairs trading technique.Findings:The five stock indices are found to be cointegrated. Analysis shows that the cointegration rank among the variables is significantly influenced by structural breaks. Two pairs of stock variables are also found to be cointegrated. This guaranteed the mean reversion property necessary for the successful execution of the pairs trading technique. Determining the optimal spread threshold also proved to be highly significant with respect to the success of this trading technique.Value:This research seeks to expand on the literature covering long-run co-movements of the volatile emerging market indices. Based on the cointegration relation shared by the BRICS, the research also seeks to encourage risk taking when investing. We achieve this by showing the potential rewards that can be realized through employing appropriate statistical arbitrage trading techniques in these markets.

In this thesis we investigate some properties of market making and statistical arbitrage applied to High Frequency Trading (HFT). Using the Hamilton-Jacobi-Bellman(HJB) model developed by Guilbaud, Fabien and Pham, Huyen in 2012, we studied how market making works to obtain optimal strategy during limit order and market order. Also we develop the best investment strategy through Moving Average, Exponential Moving Average, Relative Strength Index, Sharpe Ratio.

This thesis analyzes the performance and process of constructing portfolios of cryptocurrency pairs based on cointegrated relationships indicated by the Augmented Dickey-Fuller test, Johansen’s test and Phillips Peron’s test. Pairs are tested for cointegration over a 3-month and a 6-month window and then traded over a trading window of the same length. The cryptocurrencies included in the study are 14 cryptocurrencies with the highest market capitalization on April 24th 2019. One trading strategy has been applied on every portfolio following the 3-month and the 6-month methodology with thresholds at 1.75 and stop-losses at 4 standard deviations. The performance of each portfolio is compared with their corresponding buy and hold benchmark. All portfolios outperformed their buy and hold benchmark, with and without transaction costs set to 2%. Following the 3-month methodology was superior to the 6- month method and the portfolios formed through Phillips Peron’s test had the highest return for both window methods.

Este trabalho tem como intuito aplicar quatro estratégias de arbitragem estatística ao mercado acionário brasileiro no período compreendido entre 2004 e 2012. A primeira delas explora o fenômeno de momentum e tem como referência Jegadeesh e Titman (1993). A segunda trata de replicação de benchmarks utilizando técnicas de cointegração e foi baseada parcialmente em Alexander e Dimitriu (2002). A terceira é uma estratégia do tipo pair trade e tem referência em Gatev et al (2006). A última é uma estratégia de reversão de preços relativos de uma cesta de ações utilizando a abordagem de componentes principais e tem como referência Avellaneda e Lee (2010). Foram implementadas algumas modificações nas estratégias de forma que estas se adaptassem a certas características do mercado brasileiro e possuíssem maior grau de realismo: 1) o nível de alavancagem foi controlado de forma mais rigorosa; 2) os principais parâmetros foram determinados endogenamente, com dados fora da amostra. 3) só foram consideradas ações com grau razoável de liquidez; e 4) os custos de transação foram obtidos de séries cotadas, ao invés de fixados arbitrariamente. Os resultados mostraram que, após a contabilização dos custos de transação, nenhuma estratégia foi capaz de gerar excessos de retorno positivos com alta significância estatística. O trabalho também contribui ao fornecer alguma referência de custos médios de bid ask spread e taxas de empréstimo de ações para estratégias de arbitragem estatística no mercado acionário brasileiro. / This work tests four statistical arbitrage strategies in the Brazilian stock market from 2004 to 2012. The first one explores momentum and the reference is Jegadeesh and Titman (1993). The second replicates a positive benchmark using cointegration techniques and was based, partially, on Alexander and Dimitriu (2002). The third is a pair trade strategy and is connected to Gatev et al (2006). The last one is a portfolio price reversal strategy that uses principal component analysis and is based on Avellaneda and Lee (2010). Some modifications were made in the strategies in order to adapt them to the local market characteristics and to make them more realistic: 1) the leverage was controlled more strictly; 2) the main parameters were determined endogenously; 3) Only liquid stocks were considered; 4) the transaction costs were obtained from quoted data, instead of determined arbitrarily. The results indicated that, after the transaction costs being taken in to account, no strategy was capable of generate excess returns with high statistical significance. Other contribution of this work is to give some reference of transaction costs involved in statistical arbitrage strategies in the Brazilian stock market.

In this thesis we study the problem of forecasting the final score of a football match before the game kicks off (pre-match) and show how the derived models can be used to make profit in an algorithmic trading (betting) strategy. The thesis consists of two main parts. The first part discusses the database and a new class of counting processes. The second part describes the football forecasting models. The data part discusses the details of the design, specification and data collection of a comprehensive database containing extensive information on match results and events, players' skills and attributes and betting market prices. The database was created using state of the art web-scraping, text-processing and data-mining techniques. At the time of writing, we have collected data on all games played in the five major European leagues since the 2009-2010 season and on more than 7000 players. The statistical modelling part discusses forecasting models based on a new generation of counting process with flexible inter-arrival time distributions. Several different methods for fast computation of the associated probabilities are derived and compared. The proposed algorithms are implemented in a contributed R package Countr available from the Comprehensive R Archive Network. One of these flexible count distributions, the Weibull count distribution, was used to derive our first forecasting model. Its predictive ability is compared to the models previously suggested in the literature and tested in an algorithmic trading (betting) strategy. The model developed has been shown to perform rather well compared to its competitors. Our second forecasting model uses the same statistical distribution but models the attack and defence strengths of each team at the players level rather than at a team level, as is systematically done in the literature. For this model we make heavy use of the data on the players' attributes discussed in the data part of the thesis. Not only does this model turn out to have a higher predictive power but it also allows us to answer important questions about the 'nature of the game' such as the contribution of the full-backs to the attacking efforts or where would a new team finish in the Premier League.

Stationarity is an essential property to portfolio return in the past statistical arbitrage strategy, this article uses Neo-75 rule, momentum effect, properties as independent and identically distribution and stationarity in error term, in one asset and in the very short holding period. The result in out sample period owning positive cumulative return. The finding suggests individual investors use this strategy in higher efficiency market to avoid invalidation in our model. This article surveyed CAC40, DJI, HangSeng, NASDAQ, Nikkei225, Shanghai and TWII indices. All the excess returns in out sample periods indicate they are exclude weak form of efficient market.

Este trabalho tem como intuito aplicar quatro estratégias de arbitragem estatística ao mercado acionário brasileiro no período compreendido entre 2004 e 2012. A primeira delas explora o fenômeno de momentum e tem como referência Jegadeesh e Titman (1993). A segunda trata de replicação de benchmarks utilizando técnicas de cointegração e foi baseada parcialmente em Alexander e Dimitriu (2002). A terceira é uma estratégia do tipo pair trade e tem referência em Gatev et al (2006). A última é uma estratégia de reversão de preços relativos de uma cesta de ações utilizando a abordagem de componentes principais e tem como referência Avellaneda e Lee (2010). Foram implementadas algumas modificações nas estratégias de forma que estas se adaptassem a certas características do mercado brasileiro e possuíssem maior grau de realismo: 1) o nível de alavancagem foi controlado de forma mais rigorosa; 2) os principais parâmetros foram determinados endogenamente, com dados fora da amostra. 3) só foram consideradas ações com grau razoável de liquidez; e 4) os custos de transação foram obtidos de séries cotadas, ao invés de fixados arbitrariamente. Os resultados mostraram que, após a contabilização dos custos de transação, nenhuma estratégia foi capaz de gerar excessos de retorno positivos com alta significância estatística. O trabalho também contribui ao fornecer alguma referência de custos médios de bid ask spread e taxas de empréstimo de ações para estratégias de arbitragem estatística no mercado acionário brasileiro. / This work tests four statistical arbitrage strategies in the Brazilian stock market from 2004 to 2012. The first one explores momentum and the reference is Jegadeesh and Titman (1993). The second replicates a positive benchmark using cointegration techniques and was based, partially, on Alexander and Dimitriu (2002). The third is a pair trade strategy and is connected to Gatev et al (2006). The last one is a portfolio price reversal strategy that uses principal component analysis and is based on Avellaneda and Lee (2010). Some modifications were made in the strategies in order to adapt them to the local market characteristics and to make them more realistic: 1) the leverage was controlled more strictly; 2) the main parameters were determined endogenously; 3) Only liquid stocks were considered; 4) the transaction costs were obtained from quoted data, instead of determined arbitrarily. The results indicated that, after the transaction costs being taken in to account, no strategy was capable of generate excess returns with high statistical significance. Other contribution of this work is to give some reference of transaction costs involved in statistical arbitrage strategies in the Brazilian stock market.

This thesis deals with statistical arbitrage as a strategy applied in algorithmic trading of US Treasury bonds in the selected timeframe from 1980 until 2017. Our aim is to prove that a specific event on the treasury market, namely reopening of the bonds, constitutes an arbitrage opportunity that enables the investor to systematically yield extraordinary profits on the market. This thesis includes a theoretical introduction to algorithmic trading and statistical arbitrage. Based on this introduction we formulate hypotheses, which are then tested in the application part by constructing an algorithm that simulates a trading strategy on historical data. Comparing three strategies we determined that this strategy is meaningful, or performs better than a random walk and that it is profitable.

The study describes how cointegration-based techniques can be employed in order to construct profitable trading strategies that exploit mispricing events between similar securities. Particularly, the Johansen Maximum Likelihood Estimation and the Kalman filter approaches are applied to the universe of 200 most liquid ETF stocks traded on NYSE and NASDAQ. The results show that the strategies are quite sensitive to transaction costs, but are still able to maintain profitability even after accounting for a conservative level of transaction costs. While the Kalman filter produces better results on daily data, the 15-minute timeframe is dominated by portfolios constructed by the Johansen cointegration test. Both strategies achieve significantly higher risk-adjusted returns on the intraday timeframe. The study also reveals a performance decline of both strategies in the period of 2013-2015 and outlines possible interpretation of such event.