3790 research outputs found
Sort by
Comparative Analysis of Financial Performance Among Traditional Quantitative, AI-Based, and Hybrid Modeling Approaches
No full textThis work examines the evolution and integration of traditional quantitative models with artificial intelligence (AI) and machine learning (ML) techniques in financial forecasting. It highlights the strengths and limitations of classical statistical approaches such as ARIMA and GARCH, emphasizing their interpretability and stability under consistent market conditions, while acknowledging their challenges in capturing nonlinear dynamics and adapting to structural shifts. The study explores advanced AI architectures, including LSTM networks, CNNs, and attention mechanisms, which offer enhanced capabilities for modeling complex temporal and spatial patterns in heterogeneous financial data. Hybrid modeling strategies that combine statistical rigor with AI adaptability are analyzed, demonstrating improvements in predictive accuracy, noise reduction, and risk-adjusted performance across various asset classes and market domains. The integration methodologies, advantages, and potential drawbacks of such hybrids are discussed, alongside comprehensive evaluation frameworks employing return-based, risk-adjusted, error, and directional accuracy metrics. Broader applications in stock markets, foreign exchange, cryptocurrencies, derivatives, and diverse trading strategies are reviewed, with attention to uncertainty quantification and its impact on model reliability. The importance of explainable AI techniques and transparency in model outputs is underscored to meet regulatory and operational requirements. Human-machine collaboration is presented as a means to combine computational power with expert judgment effectively. Finally, considerations of computational complexity, scalability, ethical implications including data privacy, fairness, and regulatory compliance are addressed, highlighting current trends and future challenges in developing financial forecasting systems that balance accuracy, interpretability, and operational feasibility.
Keywords: Financial Forecasting, Hybrid Models, Deep Learning, Algorithmic Trading, Quantitative Finance, Sentiment Analysis, Model Interpretability, Risk-Adjusted Performance, Time-Series Analysis, Explainable AI (XAI)n
Artificial Intelligence Applications in Financial Markets and Corporate Finance: Technologies, Challenges, and Opportunities
No full textThis study examines the transformative impact of artificial intelligence (AI) on financial markets and corporate finance, highlighting its role in enhancing analytical precision, operational efficiency, and strategic decision-making. It explores the historical evolution of AI integration, from early automation to advanced machine learning and deep learning applications, emphasizing their contributions to market analysis, risk management, and portfolio optimization. The paper discusses key AI techniques, including natural language processing, reinforcement learning, and generative models, and their deployment across trading, credit assessment, and corporate governance. Attention is given to data management challenges, ethical considerations such as bias mitigation and transparency, and regulatory compliance in AI-driven financial systems. The work also addresses organizational and cultural factors influencing AI adoption, as well as the societal implications related to financial inclusion and workforce transformation. Methodological approaches encompass quantitative modeling, qualitative insights, and bibliometric analyses, providing a comprehensive overview of AI’s integration within finance. Finally, the study identifies opportunities and challenges associated with AI implementation, underscoring the need for responsible governance and continuous innovation to realize sustainable benefits in the financial sector.
Keywords: Artificial Intelligence in Finance, Financial Technology (FinTech), Algorithmic Trading, Deep Learning, Explainable AI (XAI), Regulatory Technology (RegTech), AI-driven Risk Management, Generative AI, Corporate Finance, Ethical AI in Financen
Review of Riya Das’s Women at Odds: Indifference, Antagonism, and Progress in Late Victorian Literature
No full textThis cogent work, "highly recommended" by CHOICE, is yet another example of the fine new scholarship that, fortunately, often targets George Eliot's work. One might occasionally disagree with the author, but even sceptical engagement in this case enables greater clarity in one's own thinking
Review of Priyanka Anne Jacob’s The Victorian Novel on File, Secrets, Hoards, and Information Storage
No full textThe British Library began the nineteenth century with 48,000 volumes and ended it with two million. Add to their extraordinary mound of paper the holdings of other libraries, the Public Records Office, newspapers and magazines, the records of politics, commerce, and empire, etcetera..
Photosynthetic pigment and cyanotoxin data for: Cyanotoxin production in shallow subtropical lakes is driven by nutrient enrichment and primary producer abundance on the millennial scale
No full textIncreased cyanotoxin concentrations from harmful algal blooms (HABs) in lake systems pose a global challenge to water quality. Although progress has been made in monitoring cyanotoxins in modern environments over recent decades, identifying the triggers of cyanotoxin release by cyanobacteria has yielded mixed results from experimental and analytical studies. Paleolimnological reconstructions can reveal whole-lake long-term changes, but few studies have directly measured cyanotoxins alongside other water quality proxies. Here, we investigated the drivers of sedimentary total microcystin (MC) concentrations on millennial scales in hypereutrophic Lakes Dora and Marian in central Florida, USA. We analyzed dated sediment records using paleolimnological techniques to reconstruct nutrient deposition, cyanobacteria abundance (photosynthetic pigments), and cyanotoxins (total MCs). The objective was to investigate the linkage between MC concentrations in the sediments with both biotic (cyanobacteria and other primary producers) and abiotic factors (nutrients and climate). We found that MC production occurred throughout the ∼7000-year period, progressing from periods of moderate to low, and then to high concentrations in both lakes. Statistical analyses showed that historical MC concentrations were correlated with sedimentary measurements of total phosphorus (TP), cyanobacteria abundance, and other primary producer groups, such as cryptophytes. However, there was only a minimal correspondence with climate proxies, such as charcoal and pollen, suggesting that internal nutrient cycling and human pressures were the dominant drivers of MC deposition. Our study demonstrates that cyanotoxins have occurred for millennia in both lakes with maintained relationships to nutrients and other environmental factors that existed both in historic and modern limnological conditions.PublishedYe
Evaluating the Influence of Natural and Artificial Light on Broiler Behavior Across Growth Stages
Full text linkIn recent years, the addition of windows to broiler houses to provide natural light during rearing has increased; however, the effects of natural light on bird behavior remain largely unknown. To evaluate the impact of raising broilers under natural or artificial light on bird behavior, mixed-sex Ross 708 chicks (N=720) were housed in 16 rooms (44 birds/room), with 8 rooms per light treatment and raised until 56 d of age. Chicks were randomly assigned to one of two light treatments: artificial light provided via a 5000K LED (AL), or natural light provided via 1 ft2 window and supplemented by a 5000K LED bulb (NL). Behavior frequency was assessed every 30 minutes over a 24-hour period via scan sampling on d 12, 33, and 54, for behaviors of; body shaking, drinking, dustbathing, eating, environmental pecking, foraging, frolicking, other active behaviors (jumping, flight, shuffle gait and adjusting posture), preening, resting (sitting and sleeping), running, self-scratching, standing, stretching, walking, and wing-flapping. Data were analyzed as a repeated measures ANOVA (PROC GLIMMIX, SAS 9.4) for main effects of light treatment and age, and means were compared via the Tukey-Kramer test. A p-value of <0.05 was considered statistically significant. An interaction between treatment and age was found for body shaking, frolicking, and resting. Body shaking occurred more on d 54 in NL (1±0.4), then in NL on d 12 (0) and 33 (0), and AL on d 12 (0) and 54 (0). Birds frolicked more on d 12 in NL (2.5±0.9) than on d 12 (0), 33 (0.25±0.25), and 54 (0) in AL, and d 54 in NL (0). Rest was highest on d 54 for both AL (1698±8.8) and NL (1658.5±7.5); followed by d 33 in AL (1572.25±13), then d 33 in NL (1535±25), and d 12 in NL (1472±5.7) and lowest on d 12 in AL (1430±19.4).
Results showed that NL birds performed more drinking and active behaviors (60.6±2.4 and 26.2±2.7 respectively) compared to those raised under AL (51.6±2.4 and 17.2±1.6, respectively). Running (0.6±0.2), standing (87.1±7.7), and dust bathing (7±1.3) were performed more on d 12 than on either d 33 or 54. Whilst eating (139±8.5) and walking (57.5±3.2) were performed most on d 12, then d 33, and least on d 54. Foraging was performed more on d 12 (65±10.1) and 33 (74.9±4.5) than on d 54 (40.1±2.8). Birds preened most on d 33 (73.2±2.7). Stretching was most frequent on d 33 (40.9±3.4), followed by d 12 (15.9±1.1), with the least amount observed on d 54 (6.7±0.9). Overall, this study suggests that the NL treatment influences various behaviors in birds, including nutritive, play, and other active behaviors. Birds raised under NL conditions exhibited more frequent drinking, active behaviors and frolicking compared to those raised under AL conditions.PublishedYe
Charcoal from bat guano in Cripps Mill Cave
No full textThis is charcoal data extracted from a bat guano sediment core that was collected from Cripps Mill Cave in Tennessee, US
No field? No problem! Enriching IR metadata via the DOI registrar
No full textMany DSpace institutional repositories rely on the Dublin Core metadata schema, limiting what can be described in the metadata records. This is true even when IRs locally adapt or extend Dublin Core. When IRs accept research data to help institutional personnel comply with public access mandates or journal requirements, this becomes a bigger issue. Rich metadata is crucial for data deposits not only to describe the data itself, thereby facilitating reuse, but also to ensure they are properly represented in the growing web of interconnected reporting systems that track research outputs. For example, dataset contributors should be linked to other entities through permanent identifiers such as ORCIDs, and bidirectional linkages to all related publications and software should exist. This talk describes an ongoing project at Auburn University Libraries to improve the metadata shared about data deposits in the institutional repository using its DOI registrar. As a member of DataCite, Auburn University utilizes its services to assign DOIs to items in the IR. The DSpace IR employs a modified Dublin Core metadata schema, but describing data using the more complex DataCite metadata schema would allow for much greater functionality. Since September 2023, a librarian and an undergraduate have been working together to decouple the local DC records from the DOI records in DataCite and create a process to feed more extensive metadata to the latter. This dual system permits information and relationships that cannot be represented in the repository's schema to be reported via the DOI public registration metadata. It will also enable DOIs to be reserved at the draft stage, a common request from researchers. Finally, this presentation will summarize the pros and cons of moving to a more complex workflow requiring greater manual intervention
Life history strategies are decoupled from ecomorphological convergence in two Anolis lizards
No full textConvergence is considered powerful evidence for adaptation to similar natural selection pressures. However, for many ecologically and morphologically convergent species, it remains unclear if convergence extends to reproductive strategies, which are particularly important because of their tight connection to fitness. Here, by measuring key life-history traits (e.g., reproductive status, egg size, oviposition frequency, reproductive effort) across a full annual cycle comprising both reproductive and non-reproductive seasons, we discover divergence in reproductive strategies in two Anolis lizards that are otherwise strikingly convergent in ecology, morphology, and behavior. The Cuban brown anole (A. sagrei) rapidly produces many small eggs during a concentrated summer reproductive season, while the Puerto Rican crested anole (A. cristatellus) produces comparatively fewer, larger eggs over a longer period. Thus, despite evolving highly convergent ecomorphological phenotypes and both being constrained to a single-egg clutch, these species exhibit marked divergence in life-history trade-off strategies along the fast-slow continuum. Our results indicate that ecomorphological convergence evolved uncoupled from life-history pathways in these species.AccpetedYe
Yes, use AI, but not like that! Helping student researchers navigate conflicting messages about generative AI
No full textStudent use of generative AI tools in their academic research is a growing trend, but these newer technologies have undergone little testing and are predominately unregulated in higher education contexts. Having planned and presented a one-shot workshop, “Responsible Use of AI Research Tools", two academic librarians will discuss their experiences with this workshop including its development, presentation and lessons learned