This thesis aims to apply methods of value investing into developing field of algorithmic trading. Firstly, we investigate the effect of several fundamental variables on stock returns using the fixed effects model and portfolio approach. The results confirm that size and book- to-market ratio explain some variation in stock returns that market alone do not capture. Moreover, we observe a significant positive effect of book-to-market ratio and negative effect of size on future stock returns. Secondly, we try to utilize those variables in a trading algorithm. Using the common performance evaluation tools we test several fundamentally based strategies and discover that investing into small stocks with high book-to-market ratio beats the market in the tested period between 2009 and 2015. Although we have to be careful with conclusions as our dataset has some limitations, we believe that there is a market anomaly in the testing period which may be caused by preference of technical strategies over value investing by market participants.

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.

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.

As science advances, computer technologies are developing rapidly in the past decades. The previous way of traders¡¦ yelling for orders in the house of exchange has been replaced by the Internet and computers. The trading modes of institutional investors are transforming gradually, particularly the radical changes in the US stock market for the past 5 years. The transaction volume from high frequency trading and algorithmic trading is growing dramatically per year, accounting for at least 70% in the U.S. market. And many researchers find these trading methods based on the computer programs good in increasing liquidity, reducing volatility and facilitating price discovery. By using intraday data of Taiwan stock market in 2008 to conduct empirical research, this study intends to analyze the effect of this trend on the TW stock market. Empirical results found that the greater the market capitalization, liquidity, stock volatility are, the higher the proportion of algorithmic trading will be, but which only exists in foreign institutional investors. On the other hand, the increase of the proportion of algorithmic trading can improve liquidity, meanwhile raise the volatility. The conclusion remains unchanged when applied to control the effect of financial tsunami. That means algorithmic trader¡¦s behaviors are not always positive. This result could be related to the special transaction mechanism or lower competition of algorithmic trading in Taiwan. As to trading strategy, the result found that foreign institutional investors focus on momentum strategies, whereas particular dealers act for the sake of index arbitrage or hedge. In summary, the algorithmic trader¡¦s transaction bears positive (liquidity) and negative (volatility) impact on the market at the same time. For individual investors, algorithmic trading¡¦s momentum strategy could appeal to them, but they may not make a profit from these trades, because this strategy could merely want to pull price higher and sell stock or the opposite. About regulators, algorithmic traders¡¦ behavior should be regulated partly; regulatory authorities might also consider adding the circuit mechanism similar to South Koreas¡¦, especially on the program trading. Keywords: algorithmic trading, high frequency trading, intraday, strategy, liquidity, volatility, market quality

Syfte: Högfrekvenshandel har på senare år varit ett omdiskuterat och kontroversiellt ämne. Fenomenet har genomgått omfattande granskning och åsikterna kring dess påverkan på marknaden och dess aktörer går isär. Då tidigare forskning främst genomförts på den amerikanska marknaden är syftet med den här studien att bistå med en djupare insikt kring denna typ av handel och dess avtryck på den svenska finansmarknaden. Metod: För att behandla syftet har en kvalitativ studie av högfrekvenshandel med en deduktiv ansats genomförts. Teori: Studien utgår från Rational Choice Theory, Effektiva marknadshypotesen och tidigare forskning inom ämnet. Med hjälp av det teoretiska ramverket har studien analyserat det empiriska underlaget. Relevanta aspekter har identifierats som kan förklara varför studiens respondenter har ett specifikt förhållningssätt gentemot högfrekvenshandel. Empiri: Studien består av en dokumentstudie och fyra semistrukturerade intervjuer med intressenter på den svenska finansmarknaden. Intervjuerna ämnar identifiera de olika intressenternas förhållningssätt gentemot högfrekvenshandel och dess bakomliggande orsaker. Slutsats: Studien har kommit fram till att förhållningssättet gentemot högfrekvenshandel står i relation till vilken typ av verksamhet som intressenten bedriver. Vidare kan det konstateras att tidigare forskning till stor del går att applicera på den svenska marknaden. / Purpose: In recent years, High Frequency Trading has been a widely debated and controversial topic. The phenomenon has been subject to extensive examination and the opinions regarding its effect on the financial markets are inconsistent. Previous research has foremost been conducted on the American financial market. Thus the purpose of this thesis is to contribute with deeper insight regarding this kind of trading and its impact on the Swedish financial market. Method: To address the purpose of this thesis, a qualitative study with a deductive approach has been conducted. Theory: The thesis emanates from Rational Choice Theory, The Efficient Market Hypothesis and previous research within the field. Using the theoretical framework, the thesis has analyzed the empirical data. Relevant aspects has been identified which can explain why the thesis’ respondents has a specific approach towards High Frequency Trading. Empirics: The thesis consists of a document study and four semi structured interviews with stakeholders on the Swedish financial market. Through these interviews, the thesis aims to identify the stakeholders’ different approaches towards High Frequency Trading and what might cause this particular point of view. Conclusion: The thesis can conclude that the approach towards High Frequency Trading is correlated to the type of operation conducted by the respondent. Furthermore, it can be concluded that previous research in general is applicable on the Swedish financial market.

We propose a systematic framework for designing adaptive trading strategies that minimize both the mean and the variance of the execution costs. This is achieved by diversifying risk over sequential decisions in discrete time. By incorporating previous trading performance as a state variable, the framework can dynamically adjust the risk-aversion level for future trading. This incorporation also allows the framework to solve the mean-variance problems for different risk aversion factors all at once. After developing this framework, it is then applied to solve three algorithmic trading problems. The first two are trade scheduling problems, which address how to split a large order into sequential small orders in order to best approximate a target price – in our case, either the arrival price, or the Volume-Weighed-Average-Price (VWAP). The third problem is one of optimal execution of the resulting small orders by submitting market and limit orders. Unlike the tradition in both academia and industry of treating the scheduling and order placement problems separately, our approach treats them together and solves them simultaneously. In out-of-sample tests, this unified strategy consistently outperforms strategies that treat the two problems separately.

The thesis offers a framework for trading algorithm optimization and tests statistical and economical significance of its performance on American, Swedish and Russian futures markets. The results provide strong support for proposed method, as using the presented ideas one can build an intraday trading algorithm that outperforms the market in long term.

As technological changes have revolutionized the way financials assets are traded today, algorithmic trading has grown to become a major part of the world's stock markets. This study aims to explore algorithmic trading through the eyes of different market operators. The market operators have, partly based on the stakeholder theory, been categorized into six categories, namely private investors, day traders, banks, the stock market, algorithmic developers and regulators. In this study we used a qualitative research design and 11 semistructured interviews have been conducted with the market operators about the main categories risks, volatility, liquidity and monitoring. The results contributed a broader view of algorithmic trading. Respondents saw a lot of risks with the business, but the majority did not express any serious concerns about this. Volatility and liquidity were considered to be affected in both directions, depending on context. Regarding monitoring of algorithmic trading, respondents considered it necessary, but the answers differ if the current monitoringis sufficient or not. The empirical results are partly in line with previous research.

Pair trading is a well-known strategy based on statistical arbitrage. This strategy uses a short-term deviation from the mean value of the price ratio of two highly correlated stocks from the same sector as the opportunity to open a position. When ratio returns to its mean value again, the position closes. This strategy has been used for many years and the main outcome of this thesis was to test whether this strategy can be profitable even in current market conditions. For that purpose, data ranging from 2010 to April 2017 on all stocks included in the S&P 500 index were used. It was subsequently found that a pair trading strategy generated 25x higher absolute profit in comparison to random agent. Thus, it can still be considered as a profitable strategy.

This thesis will address the problem of the parallel evaluation of algorithmic trading models based on multiple kernel support vector regression. Various approaches to parallelization of the evaluation of these models will be proposed and their suitability for highly parallel architectures, namely the Intel Xeon Phi coprocessor, will be analysed considering specifics of this coprocessor and also specifics of its programming. Based on this analysis a prototype will be implemented, and its performance will be compared to a serial and multi-core baseline pursuant to executed experiments. Powered by TCPDF (www.tcpdf.org)