As the fourth digital revolution is initiated and digitalization is becoming increasingly evident in today’s society, the concept of artificial intelligence (AI) is experiencing a boom and is continuously transforming a vast variety of industries. Previous studies have found several links between AI usage and economic benefits, such as increased efficiency and lower costs. Furthermore, such benefits have been connected to financial performance indicators such as return on assets (ROA) and stock return. Additionally, the Nordic countries are known for their flourishing technological environment and the involvement in well-known technology-oriented companies. These underlying factors shaped the interest of exploring the relationship between AI usage and financial performance, as measured by ROA, stock return and the volatility of stock returns. The idea of including these three performance indicators was to get both an internal perspective, as well as a market perspective from an investors point of view while incorporating risk. This census study was conducted by performing three multiple regression models on the companies on Nasdaq OMX Nordic, which resulted in a population of 152 companies. By gathering observations between the years 2015-2019, the total number of observations amounted to 721 for the ROA model, 720 for the stock return model, and 714 for the risk model. The study follows a quantitative research design, with an objective and positivist view in regard to the research philosophical assumptions. Furthermore, a deductive research approach is taken, since previous studies, as well as theories such as the stakeholder and shareholder theories, the disruption theory, the resource-based theory and the dynamic capabilities theory are used to make conclusions. Additionally, the chosen regression model was the OLS model, incorporating the robust function since none of the regressions were fulfilling the assumption of constant variation of the error term. On a 95% confidence interval, all null hypotheses could be rejected, meaning that there was a relationship between AI usage and all performance indicators. However, the relationships were unexpectedly weak and opposing of the researchers’ expectations. As it turns out, internal performance as measured by ROA, as well as market performance measured by stock return proved to have a small negative relationship with ROA. This means that Nordic companies utilizing AI sees a negative impact on financial performance in the short run. However, risk as measured by the standard deviation (SD) of stock returns, showed a positive relationship with AI usage, meaning that investing in companies using AI is riskier. The findings contradict the idea that the economic benefits from AI cause a higher financial performance. However, since AI is just seeing a boom as of recently, it is possible that it might pay off financially in the long run.

Denna uppsats undersöker om två olika momentumstrategier applicerat på mindre bolag i Sverige kan generera överavkastning. Momentumstrategier är investeringsstrategier som bygger på historisk prisdata och enligt den effektiva marknadshypotesen ska dessa inte kunna ge en överavkastning på en svagt effektiv marknad. Flertalet studier har dock hittat bevis för att dessa fungerar, och även att momentumavkastningen ofta är större för mindre bolag än stora bolag. Studiens urval är samtliga bolag som är noterade på NASDAQ Stockholm under tidsperioden 2007-2016. Vi följer Fama och Frenchs (2008) storleksdefinitioner och sorterar in bolagen i tre kategorier efter deras börsvärde (stora, små och mikrobolag) och vi bygger sedan portföljer med bolagen som kategoriseras som små. Portföljernas avkastning riskjusteras med Fama och Frenchs (1993) trefaktormodell. Studien finner att ingen av momentumstrategierna genererar en statistiskt signifikant överavkastning på den svenska marknaden under perioden.

Aktieåterköp är ett fenomen som indikerar undervärdering av ett bolag och har historiskt lett till en positiv avvikelseavkastning vid offentliggörandet. Diskretionära periodiseringars påverkan vid offentliggörandet av aktieåterköp är inte lika undersökt. I denna uppsats undersöks diskretionära periodiseringars påverkan på avvikelseavkastningen vid offentliggörandet av aktieåterköp på den svenska marknaden från 2010 till 2019. Urvalet består av 25 observationer för att bestämma diskretionära periodiseringars roll vid den initiala marknadsreaktionen. Resultatet för studien visar en positiv avvikelseavkastning vid offentliggörandet av aktieåterköp vilket går i linje med tidigare forskning. Regressionsanalysen visar ingen kortsiktig effekt vid offentliggörandet av diskretionära periodiseringar. Slutsatsen är att diskretionära periodiseringar inte har en signifikant kortsiktig effekt på avvikelseavkastningen vid offentliggörande av aktieåterköp på den svenska marknaden under perioden.

There has been much debate among investors on the benefits cryptocurrencies can have for portfolios and how their prices moves in the market. It is not difficult to see that cryptocurrencies are very volatile, yet that does not prevent investors from pouring tons of money in crypto-investments that either generate huge returns or catastrophic losses. One of the main challenges with is cryptocurrenciesis determining how they move with the rest of the market with assets such as stocks. The objective of this thesis was to investigate whether or not crypto provides some diversification benefit and if individual cryptocurrencies move in the same manner with respect to eachother. Of special interest was if there is a relationship between the cryptocurrency market and the stock market. The cryptocurrencies chosen for this project were compared mostly to the stocks in the, very information technology-sector focused, Nasdaq 100 index along with a few other assets. This thesis was written in cooperation with Origin Group AB, an Umeå based startup firm specializing in development of cryptocurrency-related technologies, most notably blockchain. All data used comes from publicly available sources and mostly include prices for cryptocurrencies and stocks from which the daily and weekly returns were calculated. The main methods used for this thesis was four different portfolio strategies with different combinations of assets, Style analysis, and principal component analysis. The portfolio strategies showed some promise with varying tradeoffs between diversification and Sharpe-ratio but the results are a bit questionable due to the short investment period. The principal component analysis showed that the cryptocurrency price data is very noise and the currencies moves pretty much in unison in contrast to the industry sector divided Nasdaq 100, which seem to have a few more distinct directions of movement. The Style-analysis’ inconclusive results show signs of a very noisy dataset and that there may not be a clear linear relationship between conventional asset returns and those of crypto.

This paper explores how the five largest trading venues in the Nordic region compete after theimplementation of MiFID in November 2007. I investigate: (1) if NASDAQ OMX’s market sharehas increased post the introduction of major changes to its market structure, and (2) how anexchange operator can attract equity share order flow in the near future. By applying event studiesto NASDAQ OMX’s market share over time, I find that introducing a faster trading system andadmitting a high frequency trading firm as a member both have a negative impact on NASDAQOMX’s market share. The reductions in market share can be explained by high frequency tradingfirms’ trading behaviour. Introducing central counterparty clearing has a positive effect onNASDAQ OMX’s market share, which highlights market participants’ appreciation of a securetrading environment, and confirms that it is not only posting the best bid and ask quotes thatattracts order flow to an exchange. It can be concluded that NASDAQ OMX in the future needs toaddress an important trade-off between total turnover and market share, as the two are not alwayspositively correlated.

BlackRocks VD hävdar att företagens ansvar i klimatfrågan är av större vikt än någonsin, där deras prestationer inte enbart innebär en klimatrisk utan även en investeringsrisk. Denna studie undersöker huruvida företags CSR-prestation, där ESG-poäng används som proxy, påverkar marknadsreaktionen vid annonsering av vinstvarningar på den svenska marknaden under perioden 2015-2019. Med hjälp av en eventstudie baserad på 53 observationer beräknas den genomsnittliga avvikelseavkastningen. En multipel regressionsanalys genomförs för att undersöka om företagens CSR-prestation påverkar marknadsreaktionen efter att företagen utfärdar vinstvarningar. Resultatet visar att även om det sker en signifikant negativ genomsnittlig avvikelseavkastning vid annonsering av vinstvarningar, finns inget signifikant samband mellan marknadsreaktionen och företagens ESG-poäng. Det är därför inte möjligt att fastställa om företag på den svenska marknaden med en bättre ESG-prestation uppvisar mindre negativ avvikelseavkastning efter att de annonserat en vinstvarning.

Enligt tidigare forskning påverkar fusioner och förvärv inte enbart de direkt inblandade parterna, utan det har även visats medföra olika signalvärden till branschen i stort. Genom en eventstudie omfattande 23 uppköpsbud och 79 konkurrenter på Nasdaq Stockholm undersöker denna studie om det föreligger positiv abnormal avkastning för konkurrenter till budmottagande företag samt olika variablers påverkan under åren 2014–2019. Utifrån studiens resultat kan det inte påvisas statistiskt signifikant positiv abnormal avkastning bland konkurrenter till budmottagande företag, vilket kan bero på problematiken i att identifiera direkt jämförbara konkurrerande företag. Vidare kan valet av marknad, tidsperiod och dataurval även påverkat studiens resultat. Studien kan däremot påvisa att mindre konkurrenter i förhållande till målföretaget upplever en högre positiv abnormal avkastning medan uppköpsbud från utländska bolag påverkar konkurrenternas abnormala avkastning negativt. Resultatet är intressant och något förvånande då sambandet inte kunnat påvisats i tidigare studier och går delvis emot de rådande teorierna inom området.

We consider the problems commonly encountered in asset management such as optimal execution, portfolio construction, and trading strategy implementation. These problems are generally difficult in practice, in large part due to the uncertainties in financial markets. In this thesis, we develop data-driven approaches via machine learning to better address these problems and improve decision making in financial markets. Machine learning refers to a class of statistical methods that capture patterns in data. Conventional methods, such as regression, have been widely used in finance for many decades. In some cases, these methods have become important building blocks for many fundamental theories in empirical financial studies. However, newer methods such as tree-based models and neural networks remain elusive in financial literature, and their usabilities in finance are still poorly understood. The objective of this thesis is to understand the various tradeoffs these newer machine learning methods bring, and to what extent they can improve a market participant’s utility. In the first part of this thesis, we consider the decision between the use of market orders and limit orders. This is an important question in practical optimal trading problems. A key ingredient in making this decision is understanding the uncertainty of the execution of a limit order, that is, the fill probability or the probability that an order will be executed within a certain time horizon. Equivalently, one can estimate the distribution of the time-to-fill. We propose a data-driven approach based on a recurrent neural network to estimate the distribution of time-to-fill for a limit order conditional on the current market conditions. Using a historical data set, we demonstrate the superiority of this approach to several benchmark techniques. This approach also leads to significant cost reduction while implementing a trading strategy in a prototypical trading problem. In the second part of the thesis, we formulate a high-frequency optimal execution problem as an optimal stopping problem. Through reinforcement learning, we develop a data-driven approach that incorporates price predictabilities and limit order book dynamics. A deep neural network is used to represent continuation values. Our approach outperforms benchmark methods including a supervised learning method based on price prediction. With a historic NASDAQ ITCH data set, we empirically demonstrate a significant cost reduction. Various tradeoffs between Temporal Difference learning and Monte Carlo method are also discussed. Another interesting insight is the existence of a certain universality across stocks — the patterns learned from trading one stock can be generalized to another stock. In the last part of the thesis, we consider the problem of estimating the covariance matrix of high-dimensional asset return. One of the conventional methods is through the use of linear factor models and their principal component analysis estimation. In this chapter, we generalize linear factor models to a general framework of nonlinear factor models using variational autoencoders. We show that linear factor models are equivalent to a class of linear variational autoencoders. Further- more, nonlinear variational autoencoders can be viewed as an extension to linear factor models by relaxing the linearity assumption. An application of covariance estimation is to construct minimum variance portfolio. Through numerical experiments, we demonstrate that variational autoencoder improves upon linear factor models and leads to a more superior minimum variance portfolio.

Hållbarhetsfrågor blir allt mer angelägna i dagens samhälle samtidigt som samhället och dess intressenter ställer höga krav på företag att tillämpa långsiktiga hållbarhetsstrategier. Till följd av det ökade fokuset på den sociala aspekten inom företag har debatten kring kvinnliga styrelseledamöter blivit alltmer omtalad, vilket lett till att företag numera eftersträvar att skapa en balanserad könsrepresentation i styrelser. Denna kvantitativa uppsats syftar till att undersöka om det finns ett positivt samband mellan andelen kvinnliga styrelseledamöter och företags hållbarhetsprestation i företag noterade på Nasdaq Stockholm. För att undersöka om ett positivt samband föreligger tillämpar uppsatsen en multipel linjär regressionsanalys. Resultatet visar att ett signifikant positivt samband råder mellan andelen kvinnliga styrelseledamöter och företags hållbarhetsprestation, där ett starkare signifikant samband förekommer när kvinnliga styrelseledamöter uppgår till minst 30 procent. Uppsatsen bidrar med empiriska upptäckter för den svenska marknaden och företag noterade på Nasdaq Stockholm.

No description available.