31 |
Predicting the Movement Direction of OMXS30 Stock Index Using XGBoost and Sentiment AnalysisElena, Podasca January 2021 (has links)
Background. Stock market prediction is an active yet challenging research area. A lot of effort has been put in by both academia and practitioners to produce accurate stock market predictions models, in the attempt to maximize investment objectives. Tree-based ensemble machine learning methods such as XGBoost have proven successful in practice. At the same time, there is a growing trend to incorporate multiple data sources in prediction models, such as historical prices and text, in order to achieve superior forecasting performance. However, most applications and research have so far focused on the American or Asian stock markets, while the Swedish stock market has not been studied extensively from the perspective of hybrid models using both price and text derived features. Objectives. The purpose of this thesis is to investigate whether augmenting a numerical dataset based on historical prices with sentiment features extracted from financial news improves classification performance when predicting the daily price trend of the Swedish stock market index, OMXS30. Methods. A dataset of 3,517 samples between 2006 - 2020 was collected from two sources, historical prices and financial news. XGBoost was used as classifier and four different metrics were employed for model performance comparison given three complementary datasets: the dataset which contains only the sentiment feature, the dataset with only price-derived features and finally, the dataset augmented with sentiment feature extracted from financial news. Results. Results show that XGBoost has a good performance in classifying the daily trend of OMXS30 given historical price features, achieving an accuracy of 73% on the test set. A small improvement across all metrics is recorded on the test set when augmenting the numerical dataset with sentiment features extracted from financial news. Conclusions. XGBoost is a powerful ensemble method for stock market prediction, reflected in a satisfactory classification performance of the daily movement direction of OMXS30. However, augmenting the numerical input set with sentiment features extracted from text did not have a powerful impact on classification performance in this case, as the improvements across all employed metrics were small.
|
32 |
Smart Beta : en kvantitativ studie om hur tre Smart Beta-strategier presterar på den svenska aktiemarknadenGunnarsson, Simon, Haskå, Filip January 2020 (has links)
Recently, the debate on passive versus active fund management has been a major focus on the Swedish capital market. Passive management is gaining more and more market shares. However, theories and previous research show that Smart Beta strategies outperform their passive benchmark index. The Smart Beta strategy is described as a hybrid between active and passive fund management, where it takes advantage of the low management cost of passive fund management and active fund management’s ability to select. This study presents three new Smart Beta strategies based on the key ratios ROA, profit margin and gross margin. The purpose of the study is to investigate whether any of the three Smart Beta portfolios can perform better than the Swedish market based on OMXS30 from a risk-adjusted perspective. Previous studies have shown that Smart Beta portfolios outperform their benchmark index. However, this study's contributing key figures show no excess return for the investigated period on the Swedish stock market.
|
33 |
Avvikelseavkastning i samband med resultatöverraskningar på OMXS30 - En eventstudieMalmström, Oskar, Nordström, Johan January 2023 (has links)
This thesis investigates if abnormal returns exist in connection to the release of quarterly reports depending on if the presented results overperform, underperform or are in line with the analysts expectations. An event study method is applied where the market model is used. The study is performed on the companies listed on the OMXS30 index on their last eight quarterly reports, from the first quarter 2021 to the fourth quarter 2022. An event window of three days and an estimation window of 40 days are applied. Two different event studies are performed, one with the 27 of the 30 stocks on OMXS30 where all the information is available and one where one more company is excluded due to distinctly deviant results. The first study shows significantly negative abnormal returns for the quarterly reports that are in line with analyst expectations, while no significant results were found for the quarterly reports over- or underperforming. The second study shows, in addition to negative abnormal returns for the quarterly reports in line with analyst expectations, significantly positive abnormal returns for companies exceeding analyst consensus, and significantly negative abnormal returns for companies failing to meet the analyst expectations. / Denna uppsats studerar huruvida det förekommer en avvikelseavkastning i samband med kvartalsrapportsläpp beroende på om resultatet överträffar, underträffar eller presterar enligt analytikernas förväntan. I studien appliceras en eventstudiemetod där marknadsmodellen används. Studien genomförs på företagen på OMXS30 över de åtta senaste kvartalen, från Q1 2021 till Q4 2022. Ett eventfönster och ett estimeringsfönster etableras på tre respektive 40 dagar. Två eventstudier genomförs, en på de 27 av de 30 aktiekurserna på OMXS30 där all information för studien finns tillgänglig samt en där ytterligare ett företag utesluts på grund av klart avvikande resultat. Den första studien visar att en signifikant negativ avvikelseavkastning för de kvartalsrapporter som presenterar neutrala nyheter i samband med kvartalsrapportsläpp förekommer, medan inget signifikant avvikande resultat förekommer för kvartalsrapporter med positiva eller negativa resultatannonseringar. I den andra studien förekommer utöver en negativ avvikelseavkastning vid neutrala nyheter även signifikant positiv avvikelseavkastning i samband med positiva resultatöverraskningar, samt signifikant negativ avvikelseavkastning i samband med negativa resultatöverraskningar.
|
34 |
Reporäntans påverkan på aktiemarknaden under hög- och lågkonjunktur : En eventstudie om hur olika företag påverkas av förändringar i reporäntan / Discount rates effects on the stock market during booms and recessionSanneh, Fabakary, Singh, Satbir January 2015 (has links)
Undersökningens syfte är att redogöra för hur den svenska aktiemarknaden reagerarvid förändringar av reporäntan under hög- respektive lågkonjunktur. Undersökningen kommer även att granska hur olika företag baserat på företagens omsättning på Stockholmsbörsenreagerar på förändringar av reporäntan.De teorier som används i undersökningen är teori om reporäntan och den effektivamarknadshypotesen. Studien behandlar den kvantitativa metoden med deduktiva inslag. Studien tillämpar även MacKinlays eventstudie metod för finansiering och ekonomi. Undersökningens resultat består av två perioder med 14 observationer av reporäntan under högkonjunktur och 9 observationer under lågkonjunktur.Resultatet påvisade inget samband mellan en höjning av reporänta och sänkning i aktiekursen i enighet med teori. Däremot går resultatet i enighet med teori för en sänkning av reporäntanoch en uppgång i kursen. / The purpose with this study is to disclose how Swedish stock market reacts to changes in discount rate during different cycles of economy. This study will also examine how different industries react on discount rate changes made by the Swedish Central Bank. The theories used in this study are the efficient market hypothesis theory and theory about the federal funds rate. In this study we use a quantitative research method with a deductive strategy. This study also includes MacKinlays event study methodology for finance and economics. The study includes two different time intervals, where one period has 14 observations on fund rate changes during booms and 9 observations during recession. The results for booms didn’t show any correlation between a hike in discount rate and a decline in stock market. Where as in recession period, there was a correlation between a decrease in discount rate and hikes in stock market
|
35 |
Företagsvärdering och kriser : En eventstudie om coronapandemins effekt på de noterade företagens värdering på OMXS30 / Company valuation and crises : An event study about the corona pandemic's effect on the listed companies' valuation on OMXS30Väyrynen Chytiris, Ion, Andersson, Jonas January 2021 (has links)
Till följd av marknadens reaktion på viruset covid-19 föll aktieindexet OMXS30 i början av år 2020 med cirka 32 procent på en månad, för att sedan återhämta sig och uppnå värderingsnivåer som OMXS30 aldrig tidigare varit på. Vi har valt att studera coronapandemins effekt på företagsvärderingarna av de noterade företagen på OMXS30. Syftet med uppsatsen är att bredda förståelsen för hur svenska företagsvärderingar i noterade företag kan bete sig under pågående börs- och samhällskriser. Genom att uppnå syftet kommer vi kunna öka kunskapen om övervärderade tillgångspriser kan leda till att stora värden går förlorade för investerare vid en eventuell börskris. Vi jämför företagens värderingar vid den högsta punkten innan coronapandemins utbrott (2020-02-19; indexkurs 1900.28) med värdet av samma företag ett år in i coronapandemin (2021-02-19; indexkurs 2036.97). Genom medelvärdesberäkningar och Paired Sample t-Test undersöker vi variationer i medelvärdet i sex nyckeltal och fyra värderingsmultiplar. Vi finner ingen statistiskt signifikant förändring i de undersökta nyckeltalen och värderingsmultiplarna, med undantag för en statistisk signifikant reducering av företagens totala omsättning. Vårt resultat talar för att värdeökningen i OMXS30 under den studerade perioden till stor del varit driven av de större företagen i indexet med högre viktning, snarare än en bred genomgående värdeökning för alla företagen i indexet. Vidare finner vi även att företagen verkar ha reducerat kostnader och/eller erhållit coronastöd från svenska staten för att kunna bibehålla vinstnivåerna på grund av den minskade omsättningen som coronapandemin medfört. / As a result of the market’s reaction to the virus COVID-19, OMXS30 fell 32 percent at the beginning of 2020 and was able to recover itself and reach new higher levels in which OMXS30 had never been before. We have chosen to study the corona pandemic’s effect on the company valuations of the listed companies on OMXS30. The purpose of the essay is to broaden the understanding of how Swedish company valuations in listed companies can behave during a stock market crisis and crisis in society. Through the purpose, we can increase the knowledge if overvalued asset prices can lead to large values being lost to investors in the event of a stock market crisis. We compare the valuations of the companies on the highest point before the corona pandemic (2020-02-19; index point: 1900.28) with the valuations of the same companies one year into the corona pandemic (2021-02-19; index point: 2036.97). Through calculations of averages and Paired Sample t-Test, we study whether there occurred any changes in the mean of six key figures and four valuation multiples. We find no statistical significance in neither the studied key figures nor the valuation multiples except for a statistically significant decrease in the companies’ total revenues. Our result points in a direction that suggests that the increase in value of OMXS30 during the studied period was mainly driven by an increase in value of the bigger companies with higher weights in the index rather than a broader increase in value of all companies in the index. We also find that companies might have reduced their overall cost and/or received financial support from the Swedish government to be able to maintain the previous earnings levels because of the decrease in total revenue that the corona pandemic has caused.
|
36 |
Dispersion Trading: A Way to Hedge Vega Risk in Index Options / Spridningshandel: En metod för att skydda mot Vega-risk i indexoptionerIrell Fridlund, Albin, Heberlein, Johanna January 2023 (has links)
Since the introduction of derivatives to the financial markets, volatility trading has emerged as a method for investors to make money in every market condition. In parallel with introducing derivatives to the financial markets, hedging methods have emerged and are today essential instruments for the liquidity providers active in the markets. The most commonly used hedging method is delta hedging which cancels out the directional risk in the option. Hedging the vega risk with dispersion trading seems to be both a profitable and accurate hedging method. This thesis examines the effectiveness of dispersion trading for reducing the vega risk in OMXS30 options. This is investigated by backtesting a strategy based on going short OMXS30 index volatility and long volatility on a tracking portfolio with a zero net vega. This investigation aims to determine if the dispersion trading strategy can be a reliable risk management tool. It was found that vega could accurately be hedged using dispersion trading. However, when considering the bid-ask spread, the strategy did not show profitability over the simulated period. Weighting the portfolio more in favour of companies with smaller bid-ask spreads did not show improved profitability. / Sedan introduktionen av derivat på de finansiella marknaderna har volatilitetshandel dykt upp som en metod för investerare att tjäna pengar i alla marknadsförhållanden. Parallellt med introduktionen av derivat på de finansiella marknaderna har säkringsmetoder vuxit fram och är idag väsentliga instrument för de likviditetsgivare som är verksamma på marknaderna. Den vanligaste säkringsmetoden är delta säkring som tar bort den riktade risken i optionen. Att säkra vegarisken med spridningshandel tycks vara både en lönsam och pålitlig säkringsmetod. Detta examensarbete syftar till att undersöka effektiviteten av att använda en spridningshandel för att minska vegarisken i OMXS30-optioner. Metoden involverar att simulera en strategi baserad på att vara kort volatilitet i OMXS30 och lång volatilitet på en spårningsportfölj på historisk data. Genom denna undersökning strävas det efter att avgöra om strategin för spridningshandel kan vara ett tillförlitligt verktyg för riskhantering. Det visade sig att vega kunde säkras med hjälp av spridningshandel. Strategin visade sig vara lönsam under den simulerade perioden men när köp- och sälj-spreadarna i de enskilda aktieoptioner inkluderades var det inte längre lönsamt att utföra metoden. Att vikta portföljen mer till förmån för företag med mindre köp- och sälj-spread visade inte på förbättrad lönsamhet.
|
37 |
Integrated Reporting in OMXS30 companies - An Analysis of Human Capital DisclosuresSchulz, Nathalie, Burenius, Frida January 2016 (has links)
Background and Problem: Sustainability reporting is a growing interest in today’s organizations and it is essential to report on non-financial matters. Many of the existing frameworks have been criticized for being used only of symbolical reasons which is why the concept of integrated reporting and the <IR> framework have been developed. One of the cornerstones in the <IR> framework is human capital which is one of the most valuable assets in an organization. Traditionally, employee costs have only been treated as an expense and there have been limited disclosures in corporate reports. In the current business world it is instead seen as an investment in human resources. Since previous studies have shown an increase of human capital disclosures when corporate reports become integrated, integrated reporting might be the solution to this problem. Purpose: The purpose of this study is to examine if there are differences in human capital disclosures between integrated reports and separate annual and sustainability reports in companies listed at OMXS30. Delimitations: This study’s empirical examination is limited to include the companies listed at Stockholm OMX30. Only corporate reports issued for the year 2014 are treated. Methodology: For this study a self-constructed disclosure scoreboard with human capital- related items has been used to collect data from the companies’ corporate reports. Also additional information beyond the pre-determined items has been collected to extend the data collection. Empirical Results and Conclusion: The results show that human capital seems to be a subject that is relatively little reported about. The integrated reporting companies do not disclose more information compared to non-integrated reporting companies. However, the results show that integrated reporting companies seem to have a more future-oriented focus and that the disclosures are more dispersed throughout the reports. It can be concluded that company sector and size do not affect the amount or type of information.
|
38 |
The effects of analyst’s recommendations on stock prices and trade volumes : An event study on the Swedish market.Lööf, Filip, Dahlberg, Casper January 2021 (has links)
This thesis analyzes the effects of analysts’ recommendations on stock prices and trade volumes of firms listed on OMXS30 during the three-year period 2018-2020. An event study of 313 recommendations issued during the three- year period was conducted in order to calculate the abnormal returns and abnormal volumes during the event window. Our results show only one occasion respectively where buy and sell recommendations induces abnormal returns significantly different from zero. We thereby conclude that analysts’ recommendations, on average, do not impose significant abnormal returns for OMXS30-firms during the event window. A potential investment value can be found in short selling sell recommended stocks, provided that one obtains information prior to public release. However, the nature of short selling may reduce or erase this value. Our results indicates that recommendations in general, do not contain new information and that the market to an extent, acts efficient. Positive abnormal volumes significant on the 5% level are found on three occasions, hence the majority are found to be insignificant. Significant abnormal volumes of 0,071% were found on the first post-event day of a recommendation, implying a small initial volume reaction. In general, however, the results do not show clear indications of a recommendation generating positive abnormal volumes.
|
39 |
Vad påverkar OMX Real Estate -Substansvärde eller OMXS30?Westin, Love January 2016 (has links)
The question in the study is to what extent the Swedish Real estate sector index acts as a follower of larger and broader stock indexes, and to what extent the index follows the asset values owned by the real estate companies in the asset market? The study is interesting for those trying to understand the development of share values in the Swedish real estate market as well as for those interesting in the “efficient market hypothesis”. The study makes an econometric analysis of the relationship between OMXS30, OMXS Real Estate PI, and the asset value of properties owned by Swedish real estate companies. Indexes are compared with Vector Autoregression (VAR) lag models, tested for dependence of GDP, the repo rent, and inflation. A Granger causality test is also performed. Despite discussed problems with reliability of some tests, the study finds that OMXS30 Granger cause OMX Real Estate PI. The study also finds that, during the period studied, OMXS30 and OMX Real Estate PI develop differently in the initial period but later form a similar path of performance. The asset values of the companies in the real estate market are more strongly correlated with OMXS30 than with their own sector index, OMXS Real Estate PI. No significant effects are found from GDP, the repo rent or inflation on OMXS30 or OMXS Real Estate PI. This may be seen as surprising but follows results from earlier studies.
|
40 |
Mechanical investing, man’s best friend or Foolish? : -A study on mechanical investment strategies on the Swedish stock marketLundberg, Max, Åkerlund, Jakob January 2021 (has links)
The aim of this study is to examine classical Dow-strategies, Dogs of the Dow and Foolish Four relative to each other and OMXS30GI in order to test if promises of substantial returns would be kept on the Swedish stock market during the period 2002-2019. Our empirical findings show no statistically significant excess-return generated by the Foolish Four-strategy over neither the Dogs of the Dow-strategy nor OMXS30GI. Furthermore, we found that the Dogs of the Dow-strategy produced a statistically significant excess-return over benchmark OMXS30GI, however excess-return does not remain after excluding years of great market turmoil.
|
Page generated in 0.0477 seconds