This project investigates the relationship between social interactions among marketparticipants and the emergence of cyclical trends in stock markets. Two models aredeveloped to capture the interaction dynamics among individuals within the market:a basic model and a model with a price mechanism. By conducting numerical examplesand systematic simulations, we examine the behavior of these models. Ourfindings reveal that the basic model does not generate cyclical trends in the stockmarket. However, the model incorporating the price mechanism demonstrates theability to create such trends.

Background: Stock ownership has been increasing in Sweden, with 2,7 million individual owners in 2022, up from 2,1 million in 2018. A trend shows that younger individuals are becoming more involved in stock ownership, while those over 40 are decreasing in numbers. Traditional finance theories, such as neoclassical finance, assume rational decision-making and advocate for diversified portfolios, but behavioral finance acknowledges the impact of psychological factors and biases on investment decisions. Evidence suggests that households tend to reduce diversification levels during stock market crashes, which may be influenced by demographic factors.   Purpose: The aim is to investigate whether the Covid-19 stock market crash influenced the portfolio allocation and asset preferences of Swedish private investors. Specifically, we examine whether there were changes in diversification levels and whether demographic factors such as gender, age, education, and portfolio wealth impacted investment behavior. The research seeks to provide a comprehensive understanding of how Swedish private investors responded to the stock market crash.   Method: We adopt a deductive approach, rooted in the positivistic philosophy. The data for our research was collected through a quantitative survey involving 232 participants. However, only 127 were used for the data analysis. Building upon prior research, seven alternative hypotheses were formulated and examined using the binary logistic model with the statistical tool SPSS and STATA.   Conclusion: Findings from this study show that 30% of participants reported an increased diversification in response to the Covid-19 stock market crash. The only demographic factor that had a significant impact on investors’ likeliness to alter their diversification levels were gender. Women were found to be more likely to increase their diversification levels in response to a stock market crash than men. There has been a shift in asset allocation preferences, with a growing preference for safer options such as mutual funds and ETFs, and a decrease in riskier assets such as stocks. However, we do not find any flight to liquidity.

Today, there is an overwhelming amount of data that is being collected when it comes to financial markets. For forecasting stock indexes, many models rely only on historical values of the index itself. One such model is the ARIMA model. Over the last decades, machine learning models have challenged the classical time series models, such as ARIMA. The purpose of this thesis is to study the ability to make predictions based solely on the historical values of an index, by using a certain subset of machine learning models: a neural network in the form of a Gated Recurrent Unit (GRU). The GRU model’s ability to predict a financial market is compared to the ability of a simple ARIMA model. The financial market that was chosen to make the comparison was the American stock index Nasdaq-100, i.e., an index of the 100 largest non-financial companies on NASDAQ. Our results indicate that GRU is unable to outperform ARIMA in predicting the Nasdaq-100 index. For the evaluation, multiple GRU models with various combinations of different hyperparameters were created. The accuracies of these models were then compared to the accuracy of an ARIMA model by applying a conventional forecast accuracy test, which showed that there were significant differences in the accuracy of the models, in favor of ARIMA.

Today, brokers within the stock market brokerage industry are having difficulties with accurately forecasting the trading volume that is conducted by their customers. This is especially a problem during periods of exceptionally high or low trading volumes. Solving this problem would lead to both monetary savings in terms of server costs and operational planning issues. This thesis uses three Machine Learning models (Random Forest Regressor, Linear Regression, and Support Vector Regression) to predict daily trading volume. In Machine Learning, features are variables that act as explanatory variables for the dependent variable, in this case, the daily trading volume. The primary focus of this study is to evaluate and analyze which types of feature categories are the most important. Therefore, this study uses a variety of features divided into five different categories (Temporal, Historical, Market, External, and Customer). The results from the models trained using each individual feature category are compared against each other. Secondly, this study also focuses on analyzing the performance of all feature categories together. A Naive model of a 20-day rolling average is used as a benchmark to evaluate the results. The findings of this study indicate that Machine Learning models perform better than the proposed Naive approach when predicting daily stock market trading volume. However, the difference is of a small nature. Further, the Historical feature category is the category that performs best and can therefore be argued to be the most important category when predicting daily trading volume. However, the results of this study are not of statistical significance. The findings of this study can be relevant to the research field and can be used in future studies to further investigate the feature importance in stock market trading volume prediction. / Idag har företag inom industrin för aktiemäklare svårigheter att på ett träffsäkert sätt prognostisera sina kunders handelsvolymer. Detta är särskilt ett problem under perioder med extremt höga eller låga volymer. Att lösa detta problem skulle leda till både monetära besparingar i form av serverkostnader, och även lösa operationella planeringsproblem. Denna studie använder tre olika maskininlärningsmodeller (Random Forest Regressor, Linear Regression, och Support Vector Regression) för att förutspå handelsvolym. Denna studie har som primärt fokus att utvärdera och analysera vilka typer av data som är av vikt i syfte att förutspå kommande daglig aktiehandelsvolym. Denna studie använder därmed en mängd olika variabler indelat i fem grupper (Tid, Historik, Marknad, Extern, Kund). Modellerna tränas individuellt med varje grupp och resultatet jämförs inbördes för att besvara studiens frågeställningar. Studien fokuserar även på att analysera resultatet av att träna modellerna på samtliga grupper tillsammans. För att utvärdera resultatet används en naiv modell med 20 dagars rullande medelvärde. Resultatet från denna studie indikerar att användning av maskininlärning presterar bättre än den använda naiva modellen, för att förutspå daglig handelsvolym på aktiemarknaden. Skillnaden i resultat är dock liten. Vidare visar studiens resultat att den grupp av variabler som presterar bäst är kategorin Historik. Därmed kan det sägas att denna grupp av variabler är den viktigaste gruppen för att förutspå daglig handelsvolym, av grupperna använda i denna studie. Det går dock inte att säga att resultaten i denna studie är signifikanta. Resultaten och slutsatserna från denna studie bidrar till forskningsområdet och resultaten kan i framtiden användas för att fortsätta undersöka vilka variabler som är av intresse när det kommer till att förutspå daglig handelsvolym på aktiemarknaden.

Investing in the Swedish stock market has over time proven to be an effective way to increase wealth. Nationally speaking, Sweden’s population is also one of the best in the world at investing their savings. Four out of five swedes invest at least some part of their private savings into mutual funds which approximately amounts to 8.4 million people. Consequently, in 2022, the aggregated amount of household wealth invested into fund shares and stocks was a staggering 3.1 trillion Swedish crowns. With such a huge interest in the stock market it is important to understand how risk-adjusted returns should be evaluated. Traditionally there has been a choice between active and passive investment strategies, depending on how the investor views the market's pricing of securities. This study investigates, using the Carhart four-factor model, how asset pricing varies over time depending on different levels of market volatility. The theories that have been used for this study are mainly the efficient market hypothesis and the adaptive market hypothesis. With these as a starting point, various asset pricing models have been tested (Carhart four-factor model & CAPM) and examined with statistical tests to produce reliable results. The results of this study can be used to draw conclusions that both theoretically and practically contribute to the expanding body of knowledge regarding factor models and Smart Beta investment strategies, specifically in the Swedish stock market. The study suggests that the Carhart four-factor is a reliable method to determine risk-adjusted returns in the Swedish stock market, mainly when it’s used during normal market conditions. It also appears that, based on the study’s observation of alpha, the dynamics of asset pricing in the Swedish stock market are more in line with the adaptive market theory rather than the efficient market theory. This insight can be used as an argument for how the Swedish stock market can be assumed to behave. In turn, this can give investors more understanding for which risk factors are considered significant during different times of market volatility, and how their risk premiums should be discounted when valuing securities. By emphasizing the importance of various risks being priced in different ways during different times of market volatility it is possible to manage the risk exposure of security portfolios in a more accurate and desirable way. Finally, it can be stated that the results are both on par with previous research that advocates and opposes factor models. The study found the effectiveness of the Carhart four-factor model in explaining the risk-adjusted returns to vary over time and that it cannot be assumed with statistical certainty to improve upon the CAPM in all market climates.

This study seeks to investigate the relationship between human development factors and domestic stock markets using a multiple linear regression model. Despite efforts to improve the model's explanatory power, the findings indicate that the model fails to confirm the research question. Nevertheless, the model uncovers a discernible trend in the dataset, albeit with limited explanatory capacity. These results highlight the complexity of the interplay between human development factors and domestic stock markets and suggest the need for further research and alternative modeling approaches to deepen the understanding of this relationship.

In this study we analyse the relationship between renewable energy, electricity prices, and the stock market. The impact from electricity prices on stock markets have previously been thoroughly analysed. However, our study evaluates if a country’s share of renewable energy in their electricity production impacts the strength and size of the relationship in question. We use data from eight countries of rather equal economical sizes but that uses very opposed energy sources. Sweden, Norway, Finland, and Latvia represent countries with high amounts of renewable energy. While Belgium, Netherlands, Poland, and Hungary constitute countries with low shares of renewable energy. By using daily data between January 2016 and December 2021, we aim to understand the relationship of electricity prices and stock market indices and the role of renewable energy in this relationship. We do this by using Johansen’s cointegration test as well as analysing the correlation between volatilities through a DCC-GARCH(1,1). We find that both tests indicate a negative correlation between the electricity and stock markets as well as for their volatilities. In addition, we find some disparities between countries depending on their share of renewable energy. The impact of electricity prices on the stock market tends to be more pronounced for countries that use larger shares of renewable energy. Finally, findings suggest that the energy source used for electricity production also constitute an important factor in the connectivity of the markets. Wind power was found to be the main cause to the larger fluctuations on the electricity market leading to stronger relationship to the stock market. While hydro power is the more stable option of renewable energy with smaller variances and large storage capacity, weakening the link between the electricity market and stock market.

This degree project investigates if a company’s stock price development can be predicted using the general opinion expressed in tweets about the company. The project starts off with the model from a previous project and then tries to improve the results using state-of-the-art neural network sentiment analysis and more tweet data. This project also attempts to perform hourly predictions along with daily predictions in order to investigate the method further. The results show a decrease in accuracy compared to the previous project. The results also indicate that the neural network sentiment analysis improves the accuracy of the stock price development when compared to the baseline model under comparable conditions. / Detta examensarbete undersöker om ett företags aktievärdesutveckling kan förutspås genom att använda sig av den generella opinionen hos tweets skrivna om företaget. Examensarbetet utgår ifrån en model från ett tidigare projekt och försöker förbättra resultaten från denna genom att använda sig av dels state-of-the-art sentimentanalys med neurala nätverk, dels mer tweet data. Examensarbetet undersöker både prognoser timvis samt dygnsvis för att undersöka metoden djupare. Resultaten tyder på en minskad träffsäkerhet jämfört med det tidigare projektet. Resultaten indikerar också att sentimentanalys med neurala nätverk förbättrar träffsäkerheten hos aktievärdesprognosen jämfört med tidigare sentimentanalysmetod givet jämförbara förutsättningar.

When a publicly traded company changes CEOs, the stock market will react in either a positive or negative way. This thesis uses multiple regression analysis to investigate which characteristics of the personal profile of the new CEO that might evoke positive or negative reactions from the stock market, both on one-day and one-year time perspectives. The mathematical results are compared to professional opinions regarding what defines an optimal CEO. The inefficiency of the financial markets and complexity of stocks make the mathematical results mostly insignificant. The only correlations found were a positive correlation for highly paid CEOs and a negative correlation for insider recruitment. The thesis concludes that an optimal CEO is defined by its leadership abilities, not by its personal profile. / När ett börsnoterat företag byter VD kommer aktiemarknaden att reagera på ett positivt eller negativt sätt. Denna uppsats använder multipel regressionsanalys för att undersöka vilka egenskaper hos den nya VD:n som kan framkalla positiva eller negativa reaktioner från aktiemarknaden, både på en dags och på ett års tid. De matematiska resultaten jämförs med professionella åsikter om vad som definierar en optimal VD. De ineffektiva egenskaperna hos den finansiella marknaden kombinerat med aktiers komplexitet gör de matematiska resultaten till stor del insignifikanta. De enda korrelationerna som hittades var en positive korrelation för högt betalda VD:ar och en negativ korrelation för internt rekryterade VD:ar. Uppsatsen drar slutsatsen att en optimal VD definieras av sina ledarskapsförmågor och inte av sin personliga bakgrund.

The intention of this thesis is to provide a primitive mathematical model for a financial market in which tradings affect the asset prices. Currently, the idea of a price-volume relationship is typically used in the form of empirical models for specific cases. Among the theoretical models that have been used in stock markets, few included the volume parameter. The thesis provides a general theoretical model with the volume parameter for the intention of a broader use. The core of the model is the correlation between trading volume and stock price, indicating that volume should be a function of the stock price and time. This function between price and time was made visible by the use of the trading volume process, also known as the Limit Order book. The development of this model may be of some use to investors, who could build their wealth process based on the dynamics of the process found through a Limit Order Book. This wealth process can help them build an optimal trading strategy design.