This thesis is dedicated to the application of machine learning methods to algorithmic trading. We take inspiration from intraday traders and implement a system that predicts future price based on candlestick patterns and technical indicators. Using forex and US stocks tick data we create multiple aggregated bar representations. From these bars we construct original features based on candlestick pattern clustering by K-Means and long-term features derived from standard technical indicators. We then setup regression and classification tasks for Extreme Gradient Boosting models. From their predictions we extract buy and sell trading signals. We perform experiments with eight different configurations over multiple assets and trading strategies using walk-forward validation. The results report Sharpe ratios and mean profits of all the combinations. We discuss the results and recommend suitable configurations. In overall our strategies outperform randomly selected strategies. Furthermore, we provide and discuss multiple opportunities for further research.

Algorithms have been used in finance since the early 2000s and accounted for 25% of the market around 2005. In this research, algorithms account for approximately 85% of the market. The challenge faced by many investors and fund managers is beating the Swedish market index OMXS30. This research investigates publicly available algorithms and their potential for implementation and modification to outperform the market. There is a lot of research done on the subject and most of the research found was mostly at a high academic level. Although few algorithms were found in the search, some algorithms that managed to beat other markets caught interest. The market data for this research was obtained from Nordnets closed API, specifically the historical price data of various financial securities. The algorithms use the historical price data to generate buy and sell signals which represents a trade. These trades were then used to calculate performance metrics such as the geometric mean and the sharpe ratio. The performance metrics are used to measure and compare performance with the OMXS30 using a quantitative method. On average, the algorithms did not perform well on the chosen securities, although some securities stood out in all cases. Beating the market is considered a difficult task, and this research reflects some of the challenges involved. The chosen method highlights the importance of the stocks the algorithms trade, emphasizing that stocks cannot be chosen randomly. Building a fully automated unsupervised trading system is challenging and requires extensive work. Some strategies tend to require human supervision to maximize returns and limit losses, while others yield low returns for low risk. / Algoritmer har använts inom finans sedan början av 2000-talet och utgjorde cirka 25% av marknaden runt 2005. När detta arbete utförs står algoritmer för cirka 85% av marknadsvolymen. Utmaningen som många investerare och fondförvaltare står inför är att slå den svenska marknadsindexet OMXS30. Detta arbete undersöker offentligt tillgängliga algoritmer och deras potential att implementeras och modifieras för att överträffa marknaden. Det finns mycket forskning gjord inom ämnet och majoriteten av denna forskning är på en hög akademisk nivå. Trots att få algoritmer hittades i sökningen, fanns det ett fåtal algoritmer som lyckats slå andra marknadsindex. Marknadsdata för denna forskning erhölls från Nordnets slutna API, specifikt historisk prisdata från olika finansiella värdepapper. Algoritmerna använder den historiska prisdatan för att generera köp- och säljsignaler. Dessa köp och säljsignaler användes sedan för att beräkna prestandamått som geometrisk medelvärde och riskjusterad avkastning. Prestandamåtten används för att mäta och jämföra prestanda med OMXS30 genom en kvantitativ metod. I genomsnitt presterade algoritmerna inte väl på de valda värdepappren, även om vissa värdepapper utmärkte sig i alla fall. Att slå marknaden anses vara en svår uppgift och denna forskning speglar några av de utmaningar som är involverade. Den valda metoden belyser vikten av de aktier som algoritmerna handlar med och betonar att aktier inte kan väljas slumpmässigt. Att bygga ett helt automatiserat obevakat handelssystem är utmanande och kräver omfattande arbete. Vissa strategier visade sig vara i behov av mänsklig övervakning för att maximera avkastningen och begränsa förluster, medan andra gav låg avkastning för låg risk.

The equity option expiration effect is a well observed phenomenon and is explained by delta hedge rebalancing and pinning risk, which makes the strike price of an option work as a magnet for the underlying price. The FX option expiration effect has not previously been explored to the same extent. In this paper the FX option expiration effect is investigated with the aim of finding out whether it provides valuable information for predicting FX rate movements. New models are created based on the concept of the option relevance coefficient that determines which options are at higher risk of being in the money or out of the money at a specified future time and thus have an attraction effect. An algorithmic trading strategy is created to evaluate these models. The new models based on the FX option expiration effect strongly outperform time series models used as benchmarks. The best results are obtained when the information about the FX option expiration effect is included as an exogenous variable in a GARCH-X model. However, despite promising and consistent results, more scientific research is required to be able to draw significant conclusions. / Effekten av aktieoptioners förfall är ett välobserverat fenomen, som kan förklaras av delta hedge-ombalansering och pinning-risk. Som följd av dessa fungerar lösenpriset för en option som en magnet för det underliggande priset. Effekten av FX-optioners förfall har tidigare inte utforskats i samma utsträckning. I denna rapport undersöks effekten av FX-optioners förfall med målet att ta reda på om den kan ge information som kan användas till prediktioner av FX-kursen. Nya modeller skapas baserat på konceptet optionsrelevanskoefficient som bestämmer huruvida optioner har en större sannolikhet att vara "in the money" eller "out of the money" vid en specificerad framtida tidpunkt och därmed har en attraktionseffekt. En algoritmisk tradingstrategi skapas för att evaluera dessa modeller. De nya modellerna baserade på effekten av FX-optioners förfall överpresterar klart jämfört med de tidsseriemodeller som användes som riktmärken. De bästa resultaten uppnåddes när informationen om effekten av FX-optioners förfall inkluderas som en exogen variabel i en GARCH-X modell. Dock, trots lovande och konsekventa resultat, behövs mer vetenskaplig forskning för att kunna dra signifikanta slutsatser.

The aim of this paper is first of all to determine whether deep learning methods can recover trading strategies based on historical price and volume data, with scarcity of real data in mind. The second aim is to evaluate the methods to generate a deep learning blueprint for strategy extraction. Trading strategies can be built on many different types of data, often combined from different areas. In this paper, we focus on trading strategies based solely on historical price and volume data to limit the scope of the problem. Combinations of different deep learning architectures and methods such as transfer- and ensemble methods were evaluated. The results clearly show that deep learning models can recover relatively complex trading strategies to some extent. Models leveraging transfer learning outperform other models when data is scarce and ensemble methods elevate performance in certain regards. / Målet med denna rapport är i första hand att ta reda på om djupinlärningsmetoder kan återskapa handlingsstragetier baserat på historiska priser och volymdata, med vetskapen att datan är begränsad. Det andra målet är att utvärdera metoder för att skapa en djupinlärningsmall för att utvinna handelsstrategier. Handelsstrategier kan vara byggda på många olika datatyper, ofta i kombination från olika områden. I denna rapport fokuserar vi på strategier som enbart är baserade på historiska priser och volymdata för att begränsa problemet. Kombinationer av olika djupinlärningsarkitekturer tillsammans med metoder som till exempel överföringsinlärning och ensembleinlärning utvärderades. Resultaten visar tydligt att djupinlärningsmodeller kan återskapa relativt komplexa handlingsstrategier. Modeller som utnyttjade överföringsinlärning presterade bättre än andra modeller när datan var begränsad och ensembleinlärning ökade prestandan ytterligare i vissa sammanhang. / Kandidatexjobb i elektroteknik 2022, KTH, Stockholm

Bakgrund: Bakgrunden presenterar digitaliseringens utveckling och jämför det med media som tidigare varit analogt. Därefter redogörs för utvecklingen av automatiserade finansiella rådgivare och de utvecklingsstadier det genomgått. I samband med utvecklingen påträffas olika aspekter hos unga småsparare som påverkar förtroendet som övergången från att använda mänskliga finansiella rådgivare till robotrådgivare. Syftet: Syftet med uppsatsen är att undersöka hur företag inom fondförsäljning och aktiemäklarbanker kan nyttja robotrådgivningstjänster bland unga småsparare. Undersökningen fokuserar främst på aspekterna informationsspridning och riskupplevelse.  Teoretisk referensram: Denna studie utgår från tre olika teorier, vilka omfattar Unified theory of acceptance and use of technology, Theory of perceived risk och Innovation Diffusion Theory. Metod: Studien utför en kvantitativ ansats genom en tvärsnittsdesign. Urvalsramen inkluderar unga vuxna som är 18 till 30 år. Datainsamlingen genomfördes via sociala medier och skolplattformar, och totalt deltog 151 unga högskolestudenter i enkätundersökningen.  Slutsats: Utifrån resultaten förekommer det en positiv korrelation mellan användningen och riskupplevelsen, vilket gör att robotrådgivare har en möjlighet att attrahera unga vuxna till segmentet.

Comparison of double auction bidding strategies for automated trading agents Bc. Daniel Vach Absctract In this work, ZIP, GDX, and AA automated bidding strategies are compared in symmetric agent-agent experiments with a variable composition of agent population. ZIPOJA, a novel strategy based on ZIP with Oja's rule extension for updating its optimal price, is introduced. Then it is showed that ZIPOJA underperforms in competition against other strategies and that it underperforms even against the original ZIP. Dominance of AA over GDX and ZIP is questioned and it is showed that it is not robust to composition changes of agent population and that the experimental setup strongly affects the results. GDX is a dominant strategy over AA in many experiments in this work in contrast to the previous literature. Some mixed strategy Nash equilibria are found and their basins of attraction are shown by dynamic analysis.

<p>Algorithmic trading is an automated process of order execution on electronic stock markets. It can be applied to a broad range of financial instruments, and it is&nbsp; characterized by a signicant investors&#39; control over the execution of his/her orders, with the principal goal of finding the right balance between costs and risk of not (fully) executing an order. As the measurement of execution performance gives information whether best execution is achieved, a signicant number of diffeerent benchmarks is&nbsp; used in practice. The most frequently used are price benchmarks, where some of them are determined before trading (Pre-trade benchmarks), some during the trading&nbsp; day (In-traday benchmarks), and some are determined after the trade (Post-trade benchmarks). The two most dominant are VWAP and Arrival Price, which is along with other pre-trade price benchmarks known as the Implementation Shortfall (IS).</p><p>We introduce Negative Selection as a posteriori measure of the execution algorithm performance. It is based on the concept of Optimal Placement, which represents the ideal order that could be executed in a given time win-dow, where the notion of ideal means that it is an order with the best execution price considering&nbsp; market &nbsp;conditions&nbsp; during the time window. Negative Selection is dened as a difference between vectors of optimal and executed orders, with vectors dened as a quantity of shares at specied price positionsin the order book. It is equal to zero when the order is optimally executed; negative if the order is not (completely) filled, and positive if the order is executed but at an unfavorable price.</p><p>Negative Selection is based on the idea to offer a new, alternative performance measure, which will enable us to find the&nbsp; optimal trajectories and construct optimal execution of an order.</p><p>The first chapter of the thesis includes a list of notation and an overview of denitions and theorems that will be used further in the thesis. Chapters 2 and 3 follow with a&nbsp; theoretical overview of concepts related to market microstructure, basic information regarding benchmarks, and theoretical background of algorithmic trading. Original results are presented in chapters 4 and 5. Chapter 4 includes a construction of optimal placement, definition and properties of Negative Selection. The results regarding the properties of a Negative Selection are given in [35]. Chapter 5 contains the theoretical background for stochastic optimization, a model of the optimal execution formulated as a stochastic optimization problem with regard to Negative Selection, as well as original work on nonmonotone line search method [31], while numerical results are in the last, 6th chapter.</p> / <p>Algoritamsko trgovanje je automatizovani proces izvr&scaron;avanja naloga na elektronskim berzama. Može se primeniti na &scaron;irok spektar nansijskih instrumenata kojima se trguje na berzi i karakteri&scaron;e ga značajna kontrola investitora nad izvr&scaron;avanjem njegovih naloga, pri čemu se teži nalaženju pravog balansa izmedu tro&scaron;ka i rizika u vezi sa izvr&scaron;enjem naloga. S ozirom da se merenjem performasi izvr&scaron;enja naloga određuje da li je postignuto najbolje izvr&scaron;enje, u praksi postoji značajan broj različitih pokazatelja. Najče&scaron;će su to pokazatelji cena, neki od njih se određuju pre trgovanja (eng. Pre-trade), neki u toku trgovanja (eng. Intraday), a neki nakon trgovanja (eng. Post-trade). Dva najdominantnija pokazatelja cena su VWAP i Arrival Price koji je zajedno sa ostalim &quot;pre-trade&quot; pokazateljima cena poznat kao Implementation shortfall (IS).</p><p>Pojam negative selekcije se uvodi kao &quot;post-trade&quot; mera performansi algoritama izvr&scaron;enja, polazeći od pojma optimalnog naloga, koji predstavlja idealni nalog koji se&nbsp; mogao izvrsiti u datom vremenskom intervalu, pri ćemu se pod pojmom &quot;idealni&quot; podrazumeva nalog kojim se postiže najbolja cena u trži&scaron;nim uslovima koji su vladali&nbsp; u toku tog vremenskog intervala. Negativna selekcija se defini&scaron;e kao razlika vektora optimalnog i izvr&scaron;enog naloga, pri čemu su vektori naloga defisani kao količine akcija na odgovarajućim pozicijama cena knjige naloga. Ona je jednaka nuli kada je nalog optimalno izvr&scaron;en; negativna, ako nalog nije (u potpunosti) izvr&scaron;en, a pozitivna ako je nalog izvr&scaron;en, ali po nepovoljnoj ceni.</p><p>Uvođenje mere negativne selekcije zasnovano je na ideji da se ponudi nova, alternativna, mera performansi i da se u odnosu na nju nađe optimalna trajektorija i konstrui&scaron;e optimalno izvr&scaron;enje naloga.</p><p>U prvom poglavlju teze dati su lista notacija kao i pregled definicija i teorema&nbsp; neophodnih za izlaganje materije. Poglavlja 2 i 3 bave se teorijskim pregledom pojmova i literature u vezi sa mikrostrukturom trži&scaron;ta, pokazateljima trgovanja i algoritamskim trgovanjem. Originalni rezultati su predstavljeni u 4. i 5. poglavlju. Poglavlje 4 sadrži konstrukciju optimalnog naloga, definiciju i osobine negativne selekcije. Teorijski i praktični rezultati u vezi sa osobinama negativna selekcije dati su u [35]. Poglavlje 5 sadrži teorijske osnove stohastičke optimizacije, definiciju modela za optimalno izvr&scaron;enje, kao i originalni rad u vezi sa metodom nemonotonog linijskog pretraživanja [31], dok 6. poglavlje sadrži empirijske rezultate.</p>

本研究利用雲端運算的技術，發展大量使用者使用的策略交易的系統。為滿足大量使用者的運算需求，本系統包括幾項特性： 1. 採用服務導向架構以充分使用雲端運算的特性。 2. 建立非同步事件控制機制以提供服務間非同步運算能力。 3. 採用集中式資料結構，提出收縮式肋骨網絡(SRN)資料結構，減少運算需求。 4. 提供基因演算模擬環境，讓使用者可以發展符合個人投資偏好的投資策略。 / In this study, we designed a algorithmic trading system for large numbers of users on a cloud computing plateform. So the main features of the algorithmic trading system have been as follows. 1. The use of Service-Oriented architecture in order to fully use the characteristics of cloud computing. 2. The establishment of asynchronous event control mechanism to provide services to non-synchronous computing power. 3. A centralized data structure, proposed Systolic Ribs Network (SRN) data structure, reducing the computing needs. 4. To provide the genetic algorithm simulation environment that allows users to develop in line with the investment strategy personal investment preferences.

本文以全球第一個演算法交易雲端平台－Quantopian進行研究，藉由平台社群討論區內公開之演算法交易策略，透過交易策略篩選和初步優化，以演算法交易策略為投資標的，搭配不同權重策略建構投資組合。權重策略部分，本文提出適用於組合式交易策略的績效指標加權 (Performance Index Weighted) 法，應用因子投資的觀念，融合排序相關性較低、不同面向之績效指標作為報酬率驅動因子，並參考Asness et al. (2013) 以因子排序作為權重計算依據，提供了簡單直覺、非最適化求解而且穩健的加權方式，更直接地將交易策略各面向績效的優劣反應在權重上。 根據數值分析，發現組合式交易策略長期而言，整體績效表現平均優於個別演算法交易策略，最小變異、績效指標加權和均等權重投資組合的風險亦明顯低於個別交易策略，且最小變異、績效指標加權和均等權重投資組合在降低投資組合風險的同時，並未犧牲過多報酬，風險調整後績效表現優於個別交易策略。而績效指標加權投資組合之年化報酬率、風險衡量和風險調整後績效表現皆優於最小變異、平均數－變異數、均等權重的加權投資組合，此種權重策略可使投資組合之夏普比率 (Sharpe ratio) 顯著提升，且投資組合的風險大幅降低，最大跌幅 (Max drawdown) 在四年半的實驗區間內降至10%以下的水準，風險調整後績效優異。 透過Quantopian社群演算法交易平台，個人投資者也能站在巨人的肩膀上學習，集合眾人的力量，憑藉量化交易創造出和機構法人一樣具有競爭力的投資組合。如Chan (2009) 所言，個人投資者也能憑藉量化交易，設計一套演算法交易策略。 / Quantopian is a crowd-sourced hedge fund which allows members on the platform to develop their own algorithmic strategies and even get capital allocations from Quantopian. In this paper, we constructed portfolios by Quantopian trading platform and proposed Performance Index Weighted method which generate consistently profit in our study. First, we filtered algorithmic trading strategies shared on the Quantopian community and improved the performance slightly. Second, we combined multiple algorithmic strategies with varied portfolio weight method, such as minimize-variance, performance index weighted, mean-variance, and equal weighed method to construct a portfolio. To elaborate, Performance Index Weighted portfolio is actually an application of factor investing, in which the portfolio weight depends on the ranking of performance index (factors), and these index measure returns, risk, and also risk-adjusted returns, which truly reflects how well the algorithmic strategy is. As a result, we used the performance index as a return driver and invested more in well-ranked strategies directly. Performance index weighted is a simple, robust, and fully intuitively way to construct a portfolio. In numerical analysis, we found that using multiple strategies to construct a portfolio could generate better performance than a single algorithm strategy on average. Moreover, the annual returns, risk measure, and risk-adjusted returns of Performance Index Weighted portfolio turn out to be better than minimize-variance portfolio, mean-variance portfolio, and equal weighted portfolio. As a result, Performance Index Weighted portfolio has significantly higher Sharpe ratio and lower Max Drawdown (lower than 10% in our out-of-sample test) than other portfolios, which shows excellent risk-adjusted performance. Most important of all, retail traders could learn more precisely by standing on the shoulders of giants through Quantopian trading platform. Also, by collecting wisdom from the crowd, we create an opportunity for retail traders to construct competitive portfolios just as institutional investors do.

The main task of the diploma thesis is an analysis of financial performance and position of Morgan Stanley during several successive periods before and during the subprime financial crisis. Through the analysis of trends in key items, it also demonstrates strong cyclicality of financial performance and position of the investment bank. The first chapter deals with history and key divisions of the Company. The following chapter generally discusses selected phenomena, which, as per the author's view, significantly influenced industry-wide record-breaking performance during the period before the subprime crisis, and which substantially determined Morgan Stanley's risk profile and performance's corrections later during the Crisis. The core part of the Thesis is conceived as an analysis of financial performance and position of Morgan Stanley during the selected periods. The analysis of pre-crisis period until 2006 in the third chapter demonstrates growth of activities lying behind the unprecedented profitability of the Institution. The following fourth chapter analyzes deteriorating financial performance during the subprime crisis and indicates crucial strategy changes, implemented by the Company at the end of 2008. Effects of the strategic changes and challenges of the future development of the Institution are discussed in the last chapter. Additionally, the Thesis includes annexes, which further deal with selected topics and their general relations to investments banks and two annexes which compare Morgan Stanley with its nearest peers during specific periods.