Publikační činnost akademických pracovníků FES / FES Research Outputs
Permanentní URI k tomuto záznamuhttps://hdl.handle.net/10195/34537
Procházet
1112 výsledky
Search Results
Článek Omezený přístup peer-reviewedpostprint How well do investor sentiment and ensemble learning predict Bitcoin prices?(Elsevier Science BV, 2023) Hájek, Petr; Hikkerova, Lubica; Sahut, Jean -MichelInvestor sentiment is widely recognized as the major determinant of cryptocurrency prices. Although earlier research has revealed the influence of investor sentiment on cryptocurrency prices, it has not yet generated cohesive empirical findings on an important question: How effective is investor sentiment in predicting cryptocurrency prices? To address this gap, we propose a novel prediction model based on the Bitcoin Misery Index, using trading data for cryptocurrency rather than judgments from individuals who are not Bitcoin investors, as well as bagged support vector regression (BSVR), to forecast Bitcoin prices. The empirical analysis is performed for the period between March 2018 and May 2022. The results of this study suggest that the addition of the sentiment index enhances the predictive performance of BSVR signifi-cantly. Moreover, the proposed prediction system, enhanced with an automatic feature selection component, outperforms state-of-the-art methods for predicting cryptocurrency for the next 30 days.Konferenční objekt Omezený přístup peer-reviewedpostprint Revisiting the Effect of R&D intensity on innovation Performance: An Instrument Variable Approach(Academic Conferences and Publishing International Limited, 2023) Darfo-Oduro, Raymond Kwame AdaneThe literature on innovation performance makes clear the important role of R&D in innovation performance. Studies that have investigated this relationship have generally assessed the effect of R&D investment on innovation performance. It is important to note that innovation performance and R&D investment are known to be closely associated with the possibility of reverse causality. Regressing innovation performance on R&D expenditure therefore poses an important statistical challenge of endogeneity. In the presence of endogeneity regression parameter estimates are biased and inconsistent and therefore hypothesis testing may be misleading. Using data from different sectors of OECD economies sourced from the OECD data base, instrumental variable analysis is conducted through a two stage least square using the number of R&D personnel as instrument. Again, the literature assumes that innovation is an increasing function of R&D. However, considering R&D is combined with other factors, some of which are fixed, the possibility that R&D will experience diminishing returns cannot be overlooked. This means that the dominant linear relationship authors specify in these studies may not be wholly accurate. We captured this effect by modeling a quadratic relationship to reflect the diminishing returns to innovation performance. The findings of the study show that the number of R&D persons exhibit a nonlinear inverted U-shaped relationship with innovation performance. The policy implication of the findings of the study is that R&D activities must be commensurate with the size of other organizational factors to ensure that changes in R&D activities solicit a favourable response from innovation performance.Článek Omezený přístup peer-reviewedpostprint Feature extraction from unstructured texts as a combination of the morphological and the syntactic analysis and its usage in fake news classification tasks(Springer, 2023) Szabo Nagy, Kitti; Kapusta, Jozef; Munk, MichalIn this paper, a new technique of feature extraction is proposed, which is considered an essential part of natural language processing. Feature extraction is the process of transformation of the unstructured text to a format which is recognizable by computers. This means a transformation to a vector of numbers. The study evaluates and compares the performance of three methods: M1, which is the baseline method TfIdf; M2, which combines TfIdf with POS tags; and M3, a novel technique called MDgwPosF that incorporates weighted TfIdf values based on word depths and the relative frequency of POS tags. The primary focus of the study is to assess and compare the performance of these methods, with particular emphasis on evaluating how M3 performs in comparison with M1 and M2. Two different datasets and feed-forward, LSTM and GRU neural networks were used in this study. The results showed that the feed-forward model with the proposed method MDgwPosF in moderate topology achieved the best performance across various measures. The dataset created automatically performed better than the manual dataset. The differences between methods and topologies were not statistically significant. Statistically significant differences between the classification models were proven. The MDgwPosF method achieved higher accuracy compared to the baseline TfIdf, indicating that incorporating additional information into the vector can enhance the performance of TfIdf.Článek Omezený přístup peer-reviewedpostprint Exploring the linkage between workplace precaution measures, covid-19 fear and job performance: The moderating effect of academic competence(Springer, 2023) Chaudhry, Naveed Iqbal; Rasool, Samma Faiz; Raza, Masooma; Mohelska, Hana; Rehman, Fazal urThe coronavirus disease 2019 (Covid-19) has created emerging mental health challenges around the world. Like other countries, Pakistan is also confronted with covid-19 calamities. The aim of the study is to examine the impacts of workplace measures (WM) on the job performance (JP) and covid-19 fears (CF) along with the moderating role of academic competence (AC) based on the organizational support theory (OST) and job demand and resources (JDR) theory. A quantitative approach was applied to collect data from 333 banking employees in Gujranwala (Pakistan), and the hypothesis were tested via structural equation modeling using SPSS and AMOS. The study findings indicate that workplace measures significantly affect covid-19 fears except for individual preventive measures (IPM). Similarly, workplace measures significantly impact on job performance apart from information about the pandemic (IAP). Moreover, academic competence insignificantly moderates between workplace measures and covid-19 fears, but a significant moderation is found between information about pandemic (IAP) and covid-19 fears. Meanwhile, academic competence significantly moderates between workplace measures and job performance instead of a relationship between information about pandemic and job performance. However, this study was limited only to the banking sector of Pakistan. So, it will open the doors for future researchers to investigate other cultural contexts and sectors. This research gives a holistic understanding of workplace measures and contributes to the body of knowledge by enlightening the moderating role of academic competence in the banking sector of Pakistan. These useful insights would enable practitioners and policy makers to develop more efficient strategies and workplace measures to enhance job performance and reduce the fears of covid-19 among employees.Konferenční objekt Omezený přístup peer-reviewedpostprint Social media and online behaviour among high school and university students(Slezská univerzita v Opavě, 2023) Kantorová, Kateřina; Klepek, MartinThe purpose of this paper is to describe the behavior of young people in the social media environment. The method chosen for data collection was a questionnaire survey on a representative sample of 1,706 valid respondents from Czech Republic. We used chi-square, analysis of variance, and factor analysis to explain relationships in data. Besides social media popularity ranking, our results showed that the more time respondents spend on the internet, the more social media brands they use. We also inspected the applicability of duplication of a purchase law in the social media context. Data showed that popular modes of communication among students include photos and texts followed by a video. Additionally, we found a similar use of social media among students and their parents. Last, with the use of exploratory factor analysis, we concluded that social media varies according to the content users go to it for, which carries implications for marketing communication strategy.Konferenční objekt Omezený přístup peer-reviewedpostprint Selected Multimedia Tools as a Tool to Compare Changes in the Landscape(IEEE (Institute of Electrical and Electronics Engineers), 2023) Krátký, Martin; Sedlák, Pavel; Komárková, Jitka; Jech, JakubRemote sensing is one of the most widely used methods of acquiring spatial data describing the Earth's surface, which is associated with subsequent data analysis and processing. Most often, the imagery was obtained using an aeroplane or satellite as a sensor carrier, but due to technological advances, unmanned aerial vehicles (UAVs) are increasingly used. UAVs provide very-high spatial resolution data. Multimedia tools are essential during the whole processing and visualization procedure. The paper demonstrates that UAV-borne RGB data can be successfully used to detect water surface, bare land and vegetation as key land cover types. Both supervised and unsupervised classification methods are used.Článek Omezený přístup peer-reviewedpostprint Determinants of Technological and non-technological Innovations: Evidence from Ghana' Manufacturing and Service Sectors(Routledge, 2023) Odei, Samuel Amponsah; Odei, Michael Amponsah; Toseafa, EvelynThis paper aims at investigating the various factors driving technological and non-technological innovations in the manufacturing and service sectors in Ghana. We argue that numerous previous studies have shown that digitalization, firms' collaborations, access to finance, engaging in research and development and certain firm characteristic such as age, size and ownership influence firms' aptitude and incentives to be innovative. However, in the context of developing countries like Ghana, we do not know whether these same determinants also have positive effects on stimulating innovations at the firm-level. Using a sample of 549 firms sourced from the World Bank Enterprises Survey conducted between 2007 and 2013, this study finds that the adoption of digitalization promotes non-technological (organizational) innovations than technological innovations. We also find that firms' innovation collaboration with consultants and universities rather exert no impact on technological and non-technological innovations. Our finding show that internal R&D enhance technological innovations and not non-technological innovations. The main practical implications of the study are that attempts to boosting firm-level and developing countries innovation potentials should mainly focus on improving internal R&D and innovation support activities, expanding access to finance, and upgrading Information and Communication Technology (ICT) infrastructure to enhance digitalization.Článek Omezený přístup peer-reviewedpostprint An optimized GRT model with blockchain digital smart contracts for power generation enterprises(Elsevier Science BV, 2023) Chai, Shanglei; Zhang, Xichun; Abedin, Mohammad Zoynul; Chen, Huizheng; Lucey, Brian; Hájek, PetrThe traditional power generation rights trading (GRT) market is faced with the problems of weak interconnection of electricity-carbon market and low security. Using smart contracts in the blockchain, the idea of establishing a weakly centralized GRT structure is proposed in this paper. The carbon emission factor was introduced to improve the GRT model, and carbon emission market is used to further stimulate the emission reduction vitality of generating units. The empirical results show that compared with the benchmark model and improved model 1, the improved GRT model proposed by us has the best emission reduction effect. The contribution of this paper is to make up for the existing research that cannot fully consider the impact of carbon peak and carbon neutralization on the GRT market, as well as the information security issues brought by big data trading on the GRT platform. This paper puts forward some policy implications for the decarbonization and green development of the electricity market advocated by the Chinese government.Konferenční objekt Omezený přístup peer-reviewedpostprint Market Restrictions of Contracting Out the Public Service at the Municipal Level(Springer Nature Switzerland AG, 2023) Mikušová Meričková, Beáta; Mališová, Danieal; Murínová, KristínaUnder contracting out as the alternative service delivery arrangement, a government retains responsibility for providing a service, but it hires private firms to produce and deliver it. The theory of contracting suggests that, provided certain conditions are met, contracting out has the potential to improve efficiency without sacrificing quality, compared to direct public production. In developed countries, contracting out can sometimes improve the performance of the public sector. In countries making the transition from planned to market-based economies, the situation is much more complicated. Outcomes of contracting out public service are determined by the quality of contract management, and also by the character of the market of contracted service. The aim of this chapter is to identify the scale of local public service contracting out and market constraints affecting its application. The study examines the experience with contracting out service of collection and disposal of municipal solid waste among 195 municipalities in Slovakia by primary research. The outcome of quantitative analysis suggests contracting out is a frequently used service delivery arrangement (88.26% of municipalities use contracting out waste collection and disposal services), however, the outcomes of contracting out can be affected by the high concentration of service market measured by indexes concentration ratio (CR) and Herfindahl–Hirshman index (HHI).Konferenční objekt Omezený přístup peer-reviewedpostprint The Effects of Demand-Side Policy on Firm Innovation: The Mediating Role of R&D Activities(Academic Conferences International Limited, 2023) Sein, Yee Yee; Prokop, ViktorToday, innovation is seen as the engine of economic growth. As a result, governments foster innovation activities in both the public and private sectors by using policy tools such as measures, regulations, and incentives. There are two main approaches in government innovation policy: supply-side policy and demand-side policy. Supply-side policy instruments primarily stimulate technological development and diffuse innovation capabilities. On the other hand, demand-side policy supports specific demand and minimizes technological and trade uncertainty at all stages of production. Moreover, a demand-driven procurement policy is one of the main drivers of high-tech industries. At the same time, in line with the growing importance of the knowledge economy and societal concepts, knowledge or technology has become the major determinant of innovation. In this way, companies are encouraged to undertake research and development activities in order to create innovation and advance their technologies. Moreover, the government’s support for investing in research and development (R&D) activities has become an effective solution for the firm’s innovation outputs. These R&D activities are of particular importance in transition and catch-up economies such as the Czech Republic. Therefore, this article aims to analyze the impact of demand-driven policy on business innovation through R&D activities in the Czech Republic. Data from the Community Innovation Survey and partial least squares structural equation modelling are used to achieve the purpose of this article. Our findings show that demand-driven policy (procurement for innovation) significantly and positively affects firm product and process innovation through R&D activities. The findings of our research also have practical implications for firms and policymakers in the Czech Republic. These results could also apply to Central and Eastern Europe, where we can observe similar characteristics of firms, for instance, because of their innovation performance.