Publikationer från Högskolan i Jönköping
Not a member yet
26709 research outputs found
Sort by
Price Reactions to Insider Trading : Analyzing Responses to the Announcement of Insider Trading within Different Sectors in the Swedish Stock Market
No full textThis thesis investigates whether investors can earn abnormal returns by mimicking legally reported insider trading in the Swedish stock market. While prior research has shown that insiders often possess valuable information about their firms, less is known about whether outsiders without access to non-public information can benefit by following insider trades. Our study focuses on three sectors: Financial Services, Consumer Goods and Services, and Healthcare and Pharmaceuticals. We apply an event study methodology to insider transactions reported between 2021 and 2024, analyzing stock price reactions across various time windows. We examine both purchase and sale transactions, considering differences in insider position, firm size, and sector. Our findings show that abnormal returns do occur following insider trades, particularly in specific sectors and under certain conditions. For instance, outsider imitation strategies were more effective in the consumer goods and mid-cap healthcare sectors. We also found that purchase transactions tend to generate more consistent positive reactions than sales, though there are exceptions. The results suggest that, despite public disclosure, markets may not immediately absorb insider trading information, especially in contexts with higher information asymmetry. These insights contribute to the ongoing debate about market efficiency and highlight areas where outsiders may still find opportunities in publicly available data
Can Retail Investors Capture Abnormal Returns from Structured Financial News Headlines?
No full textPurpose:This paper investigates whether retail investors can achieve abnormal returns by trading stocks based on structured positive earnings headlines published by the financial news aggregator Finwire. The aim is to determine if structured financial news provides actionable signals, enabling investors to outperform the broader market. Methodology:The study employs a quantitative event-study approach, analyzing average abnormal returns (AAR) and cumulative average abnormal returns (CAAR) across a 21-day event window. Statistical significance is evaluated through parametric and non-parametric tests, with robustness checks ensuring reliability. Theoretical perspectives:The research is grounded in the Efficient Market Hypothesis (EMH) and information asymmetry theories, integrating prior literature on investor behavior and Post-Earnings Announcement Drift (PEAD) to interpret results. Empirical foundation:The analysis covers 670 structured positive earnings headlines from 114 large-cap firms listed on the Stockholm Stock Exchange, selected through rigorous data quality and liquidity criteria. Conclusion:Findings show retail investors can secure abnormal returns only immediately after structured headlines are released. However, this advantage quickly dissipates, suggesting structured financial news enhances immediate market efficiency rather than providing sustainable trading opportunities
Booked or overlooked? Investigating quality signals in Airbnb reviews : A Quantitative Study on Factors Signalling Quality in Airbnb Reviews in an Emerging Tourism Market
No full textBackground: With travel becoming more accessible, online reviews play a key role in helping consumers evaluate accommodations, especially on peer-to-peer platforms such as Airbnb. In emerging markets such as Bali, where regulatory structures are weaker and information asymmetry is higher, reviews act as important tools to assess quality. As most research focuses on hotels or mature markets, this study aims to address that gap by investigating how five quality signals in Airbnb reviews influence booking intentions in Bali. The research is grounded in signalling theory, offering insights into consumer-decisions in emerging tourism markets. Purpose: The purpose of this study is to investigate quality signals in reviews on Airbnb and their influence on booking intentions in an emerging tourism market. Method: A quantitative approach was used, employing an online survey distributed via Prolific and to friends and family to obtain empirical data. The data was analysed using multiple regression- and moderation analysis to examine the impact of quality signals on booking intentions. Conclusion: The findings show that not all quality signals hold equal importance. Volume, valence, and timeliness significantly influence booking intentions, while trust and usefulness do not. Prior Airbnb experience further moderates these effects, highlighting the role of user familiarity in interpreting review content. These insights contribute to a better understanding i of consumer decision-making in emerging tourism markets and the role of online reviews in reducing uncertainty.
From Text to Trust : An LLM Multi-Agent System with Embedding Verification for ADAS Knowledge Graph Construction
No full textThis thesis addresses the challenge of managing complex engineering knowledge in Advanced Driver Assistance Systems (ADAS) by introducing a multi-agentic system for automated Knowledge Graph (KG) construction from unstructured technical texts. This system employs an agent-inspired pipeline where specialized LLM-powered agents for information extraction and normalization collaborate with deterministic modules for validation, ensuring the semantic consistency and structural integrity of the resulting KG. A core contribution is the integration of an embedding-based "Commonsense Verifier" to assess the plausibility of extracted facts and an "Inductive Reasoner" to enrich the KG with inferred knowledge. Evaluation demonstrated the system's end-to-end functionality, showing it effectively filters implausible information and manages redundancy, thereby validating that an orchestrated system of LLM agents and deterministic checks can create a more robust and coherent knowledge base for safety-critical domains
Social exclusion in Swedish language education
No full textSvenska skolan har som uppdrag att fostra samhällsmedborgare och att undervisningen ska utgå ifrån demokratiska grunder. Svenskämnet är ett ämne som i allra högsta grad bör lyfta den demokratiska aspekten och bygga broar mellan elevernas vardag och skola. Däremot ökar det sociala utanförskapet i skolorna bland annat på grund av gängkriminalitet. I skolorna idag möter vi många elever som lever i ett socialt utanförskap, syftet med studien är att undersöka hur svensklärare inkluderar elever i socialt utanförskap i sin undervisning på högstadiet. Syftet besvaras med hjälp av följande frågor: Hur definierar svensklärarna begreppet socialt utanförskap? Samt vilka motivationsfaktorer och metoder använder svensklärarna för att inkludera elever med socialt utanförskap i svenskämnet? Studien tar avstamp i den sociokulturella teorin och det materialet som har samlat har gjort med två olika fokusgruppsintervjuer som sammanlagt består av sju svensklärare. Resultat analyseras med hjälp av en tematiskanalys. Resultat visar att svensklärarna arbetar aktivt med elever som lever i socialt utanförskap. De anser också att relation, motivation och prestation går hand i hand och är en viktig del för att lyckas med elever som lever i socialt utanförskap
Warmth or Competence:Examining Chinese Users' Attitudes to In-car Voice Assistant
No full textThis study investigates how the warmth and competence perceptions of in-car AI voice assistants, brought about by voice type (human-like voice/synthetic voice) and communication style (task-oriented/social-oriented), affect Chinese users’ attitudes. Based on the stereotype content model (SCM), the study argues that users' perceptions of the warmth (trustworthiness, friendliness and kindness) and competence (capability, intelligence, effectiveness) of voice assistants will affect overall evaluation of the system. A 2 (voice type: human-like vs. synthetic) × 2 (interaction style: social vs. taskoriented) between-subjects experiment was conducted among 184 Chinese participants. The results show that Chinese consumers have no obvious preference for in-car AI voice assistants with different voices and communication styles, and we further explain the reasons for this result considering China's cultural background. The results provide theoretical support for human-computer interaction (HCI) and SCM research and provide practical suggestions for designing more attractive and effective in-car voice assistants
The Effects of Affordability and Social Media on Swedish Parents' Fashion Purchases for Their Children
No full textThe focus of this thesis is to understand the influence of affordability and social media in the decision of Swedish parents regarding fast fashion decisions for their children and how parents experience the tension between practical constraints and sustainability. Using semi-structured interviews with ten Swedish parents, the data were collected and examined using thematic analysis and provides an insight into how parents make decisions about consumption in the complexities of digital consumer societies. In total, five themes were identified that together, capture parental decision-making: Affordability as the Organizing Logic, Selective Sustainability, Digital Complicity, Emotional Tension, and Aspirational Movement. The themes indicate how parents navigate competing priorities within practical limitations and demonstrate some agency and vulnerability in their decisions to consume. The findings establish that affordability was the most significant organizing principle for parents' decision-making, determining what parents ultimately decide upon but also determining how parents perceive different choices and how they rationalized their decision-making. Parents practice sustainability, selectively rather than holistically, and are creative in demonstrating their values under financial constraints. Parents exercise a complex relationship with social media, actively engaging with digital media while using filtering strategies to separate social media sources based on authenticity and relevance. Many parents experience emotional tension around the discrepancy between the ideals or values they espouse and their consumption behaviors creating a variety of societal justification strategies and future-based narratives for maintaining an acceptable self-image. The findings add to theoretical knowledge, extending upon Consumer Decision-Making Theory by introducing the idea of dual agency complexity, through Social Influence Theory by illustrating parents' selective receptivity to digital influence, for studies on the "attitude-behavior gap", by demonstrating how parents compartmentalize sustainability practices, and introduces the notion of "temporal ethical framing", where parents ethically frame their current choices and behaviors as trajectories of aspiration towards more sustainable purchases and consumption practices. The findings have real-world implications for fashion brands, policymakers, parent communities and parents themselves about the potential for sustained children's fashion consumption that recognizes parents' situational constraints and to better support parents working to align their consumption behaviors with their values
Beyond Automation: Enhancing Marketing Creativity in SMEs through AI
No full textBackground: As artificial intelligence (AI) becomes increasingly embedded in digital marketing, small and medium enterprises (SMEs) face both opportunities and challenges. While AI enables enhanced efficiency and data-driven strategies, concerns remain regarding its potential to replace human creativity. Previous studies have primarily emphasized automation and productivity, leaving a gap in understanding how AI might support rather than replace creative marketing processes in SMEs facing resource limitations. Purpose: The study investigates how AI can be integrated by SME marketing teams to support and enhance creativity instead of simply automating tasks. By focusing on creativity as a strategic asset, the research explores the relationship between AI tools and human innovation in digital marketing, contributing to both practical and theoretical understandings of AI’s creative potential in smaller businesses. Method: A qualitative research approach was adopted using an interpretivist paradigm. Semi-structured interviews were conducted with eight marketing professionals working in SMEs and with experience using AI tools. Thematic analysis was employed to identify patterns and insights, guided by Amabile’s Componential Theory of Creativity as the theoretical framework. Conclusion: Findings reveal that AI acts as a valuable creative assistant, supporting ideation, enhancing productivity, and enabling experimentation. However, the effectiveness of AI integration depends on thoughtful implementation, ethical transparency, and the preservation of human judgment and emotional sensitivity. The study extends Amabile’s theory by conceptualizing AI as an external environmental factor that can enhance skills, creativity related processes, and task motivation when appropriately managed. This research contributes to closing the literature gap on SME specific AI adoption and highlights the importance of balancing technological efficiency with human originality in the age of intelligent marketing tools
The Role of AI in Facilitating Startup Formation: Supporting Opportunity-Driven Entrepreneurs in Latin America
No full textArtificial Intelligence (AI) is becoming more accessible to small businesses around the world, but little is known about how it is used in low-resource settings. This study explores how AI supports opportunity-driven entrepreneurs in Guatemala as they face challenges like weak institutions, limited education, and poor access to financing. Drawing on institutional theory, experiential learning, and sociomateriality, the study uses a qualitative, interpretivist approach with semistructured interviews and additional informal conversations. The findings show that while AI isoften used for simple tasks like marketing and planning, it is not yet widely trusted or integrated into core business strategies. Entrepreneurs learn to use AI through trial and error, and their decisions are shaped by social norms, access to support, and daily business realities. The study highlights that AI alone does not transform entrepreneurship, it is most useful when supported by stronger institutions, digital education, and better infrastructure. These insights are useful for policymakers and development programs aiming to promote inclusive innovation in emerging economies
Moral Outrage and Cancel Culture in the Digital Age : A Qualitative Study of Generation Z
No full textBackground: Cancel culture involves withdrawing support from public figures or brands due to perceived wrongdoing and is often driven by social justice concerns. Central to this phenomenon is moral outrage, emotional reactions like anger and disgust, now rapidly expressed and amplified through social media, intensifying public accountability. Existing research has mainly focused on the outcomes of cancel culture, and limited research has explored the emotional and behavioural factors behind it. This thesis addresses this gap in marketing literature by examining how moral outrage shapes consumer responses to perceived moral violations. By integrating the concept of moral outrage from sociology into marketing literature, the thesis provides new insights into this gap in consumer behavior and the influence of moral outrage in the dynamics of cancel culture. Purpose: This thesis aims to explore the influence of moral outrage within cancel culture from a consumer perspective, addressing the research gap by combining empirical findings and theoretical insights to enhance academic understanding and guide future marketing practices. Method: This study employed a qualitative research design developed within the interpretivist paradigm to explore how moral outrage influences cancel culture. An abductive research approach was adopted, allowing for iterative movement between existing theories and empirical data. Data were collected through two semi-structured online focus groups, each comprising six Generation Z participants. A thematic analysis was conducted to identify latent patterns and underlying values in participants' responses, with codes developed collaboratively by the three authors. Conclusion: The results deepen understanding of moral outrage in cancel culture by revealing how social influence, performativity, and digital environments shape reactions. By examining Generation Z’s perspectives, this study offers a clearer, consumer-based understanding of how moral outrage operates and contributes to existing literature by highlighting how moral values are shaped and amplified by social media environments.