Synergy between stock prices and investor sentiment in social media

The underlying assumption of using investor sentiment to predict stock prices, stock market returns, and liquidity is that of synergy between stock prices and investor sentiment. However, this synergistic relationship has received little attention in the literature. This paper investigates the synergistic pattern between stock prices and investor sentiment using social media messages from stock market investors and natural language processing techniques. At the macro level, we reveal extremely significant positive synergy between investor sentiment and stock prices. That is, when a stock price rises, investor sentiment rises, and when a stock price falls, investor sentiment falls. However, this synergy may be reversed or even disappear over a specific time period. Through a segmented measurement of the synergy between stock prices and investor sentiment over the course of a day, we also find that investor sentiment on social media is forward looking. This provides theoretical support for using investor sentiment in stock price prediction. We also examine the effect of lockdowns, the most draconian response to COVID-19, on synergy between stock prices and investor sentiment through causal inference machine learning. Our analysis shows that external anxiety can significantly affect synergy between stock prices and investor sentiment, but this effect can promote either positive or negative synergy. This paper offers a new perspective on stock price forecasting, investor sentiment, behavioral finance, and the impact of COVID-19 on the stock markets. Copyright (c) 2022 Borsa Istanbul Anonim S, irketi. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Researchers on ice? How the COVID-19 pandemic has impacted Antarctic researchers

The COVID-19 pandemic and pandemic-related measures have impacted the lives and work-related activities of Antarctic researchers. To explore these impacts, we designed, piloted and disseminated an online survey in English, Russian, Spanish and Chinese in late 2020 and early 2021. The survey explored how the pandemic affected the productivity of Antarctic researchers, their career prospects and their mental wellbeing. Findings exposed patterns of inequities. For instance, of the 406 unique responses to the survey, women appeared to have been affected more adversely than men, especially in relation to mental health, and early-career researchers were disadvantaged more than their mid- or late-career colleagues. Overall, a third of the research participants reported at least one major negative impact from the pandemic on their mental health. Approximately half of the participants also mentioned that the COVID-19 pandemic had some positive effects, especially in terms of the advantages that working from home brought and opportunities to attend events, network or benefit from training workshops online. We conclude with a series of recommendations for science administrators and policymakers to mitigate the most serious adverse impacts of the pandemic on Antarctic research communities, with implications for other contexts where scientific activities are conducted under extreme circumstances.

Discovering AI-enabled convergences based on BERT and topic network

Various aspects of artificial intelligence (AI) have become of significant interest to academia and industry in recent times. To satisfy these academic and industrial interests, it is necessary to comprehensively investigate trends in AI-related changes of diverse areas. In this study, we identified and predicted emerging convergences with the help of AI-Associated research s collected from the SCOPUS database. The bidirectional encoder representations obtained via the transformers-based topic discovery technique were subsequently deployed to identify emerging topics related to AI. The topics discovered concern edge computing, biomedical algorithms, predictive defect maintenance, medical applications, fake news detection with block chain, explainable AI and COVID-19 applications. Their convergences were further analyzed based on the shortest path between topics to predict emerging convergences. Our findings indicated emerging AI convergences towards healthcare, manufacturing, legal applications, and marketing. These findings are expected to have policy implications for facilitating the convergences in diverse industries. Potentially, this study could contribute to the exploitation and adoption of AI-enabled convergences from a practical perspective. © 2023 Korean Society for Internet Information. All rights reserved.

The arrhythmic events and short-term cardiovascular outcomes in patients hospitalized with COVID-19 infection in Taiwan

Objective: The aim of this study is to investigate the arrhythmic events and short-term cardiovascular (CV) outcomes in patients hospitalized with COVID-19 infection in a single Taiwan tertiary center. Method(s): A retrospective study was carried out on 186 confirmed COVID-19 infection patients admitted to our hospital between May, 2021 and September, 2021. We investigate their CV symptoms, vital signs, laboratory examinations, arrhythmic events, and major adverse cardiovascular events (MACE), including ischemic stroke or systemic embolism, myocardial infarction, CV death, and heart failure (HF) during hospitalization. Result(s): During the hospitalization, 29.6% of patients had an elevation of cardiac enzymes, 67.2% had an elevation of d-dimer level, and 7.5% had abnormal NT-pro BNP level. The most common recorded arrhythmia is sinus tachycardia (22%), followed by atrial arrhythmia (12.4%, including atrial fibrillation 7.0%), sinus bradycardia (3.2%), ventricular arrhythmia (1.6%), and paroxysmal supraventricular tachycardia (1.1%). A total of 68 patients (36.6%) had arrhythmic events during hospitalization. During the mean follow-up of 2.8 months, 17 patients (9.1%) developed MACE, including 6 ischemic strokes, one pulmonary embolism, one peripheral artery occlusive disease, 3 HF, and 7 CV death. The total mortality rate is 19.9%. The hospitalized patients with arrhythmic events were associated with a higher incidence of intubation (32% vs 15%, p = 0.0062), MACE (22% vs 2%, p < 0.001), and mortality (37% vs 10%, p < 0.001) than those without arrhythmic events. Conclusion(s): The patients hospitalized with COVID-19 infection were associated with higher CV manifestations and arrhythmic events in Taiwan. Those patients with arrhythmic events were associated with higher morbidity and mortality.

Plasma ACE2 activity is increased in patients recovered from SARS-CoV-2 infection: Implications for the prolonged consequences of COVID-19

Objective: COVID-19 caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), utilises the catalytic site of membrane-bound angiotensin converting enzyme 2 (ACE2) for cell entry. It is thought that endocytosis of ACE2 results in a decrease in membrane bound ACE2 expression, and disruption of the local tissue renin angiotensin system protection. In this study, we hypothesised that SARS-CoV-2 infection would be associated with shedding of ACE2 leading to increased plasma ACE2 activity. Design and method: Australians aged >18 years (n=66) who had recovered from SARS-CoV-2 infection (positive result by PCR testing) and uninfected controls (n=70) were recruited. Serial samples were available in 23 recovered SARS-CoV-2 patients. Plasma ACE2 activity was measured using a fluorescent substrate-based assay and levels were compared using the Mann-Whitney or Kruskal-Wallis test. Serial ACE2 activity were analysed using the Friedman test for repeated measures. Post-hoc analysis was performed with a Bonferroni correction. Two-tailed P-values <0.05 were considered significant. Results: Controls and SARS-CoV-2 recovered patients were matched for age (mean±SD, 54±11 vs. 53±14 years, p=0.47) and gender (53% vs. 59% male, p=0.49). There were no significant differences (p>0.05) in the proportion of hypertension, obesity, diabetes, cardiovascular disease, or use of anti hypertensive, lipid lowering, and anti-platelet medications between the controls and SARSCoV-2 patients. Plasma ACE2 activity at median 35 days post-infection [interquartile range 30-38 days] was 97-fold higher in SARS-CoV-2 patients compared to controls (5.8 [2-11.3] vs. 0.06 [0.02-2.2] pmol/min/ml, p<0.0001). Plasma ACE2 activity was significantly different across disease severity (p=0.033), with severe COVID-19 associated with higher ACE2 activity compared to mild disease (11.2 [8.3-23.2] vs. 5.4 [1.8-9.0] pmol/min/ml, p=0.027). Men recovered from SARS-CoV-2 had higher ACE2 levels compared to women (9.2 [5.8-15.3] vs. 2.1 [0.2-5.1] pmol/min/ml, p<0.0001). In 23 patients who had serial blood samples at 63 [56-65] and 114 [111-125] days post infection, median ACE2 activity remained persistently elevated with no differences between time-points (p>0.05). Conclusions: Plasma ACE2 activity is elevated after SARS-CoV-2 infection and remains elevated post-infection. Our findings indicate the need for ongoing investigation to determine if ACE2 levels identify people at risk of prolonged illness following COVID-19.