114,698 research outputs found
The Impact of Institutional Differences on Derivatives Usage: A Comparative Study of US and Dutch Firms
Get PDFThis paper tests the influence of institutional differences on risk management practices.Several survey studies have investigated derivatives usage for risk management purposes in the US (see, among others, Bodnar, Hayt, Marston and Smithson, 1995 and Bodnar, Hayt and Marston, 1996, 1998).In this paper, we compare derivative practices of US and Dutch firms.This comparison is interesting because the institutional setting for Dutch firms differs from the US setting with respect to shareholder orientation, international trade, disclosure regulation, and the reliance on financial markets.In a number of survey studies additional countries have been studied, such as New Zealand (Berkman, Bradbury and Magan, 1997), Sweden (Alkebäck and Hagelin, 1999) and Germany (Bodnar and Gebhardt, 1999).In contrast with these papers, we facilitate a comparison by applying a matching and a weighting strategy, which corrects for different distributions over industry and size classes in the Dutch and US samples.After these corrections, the remaining results can be attributed to institutional differences.We find that Dutch firms hedge more financial risk. Because of the greater openness of the Netherlands, Dutch firms experience far more foreign exchange exposure and hedge more currency risk.US firms have more concerns regarding derivative usage, which may be linked to the stricter disclosure requirements in the US.US firms also focus more on accounting earnings, which may be attributable to the shareholder orientation in the US versus the stakeholder orientation in the Netherlands.Whereas Dutch firms tend to rely on OTC-transactions, US firms use exchange-traded derivatives and therefore require a higher counter party rating for derivatives transactions. This distinction can be accredited to the differences in the financial environments between the US and the Netherlands.The aforementioned results indicate that institutional differences between the US and the Netherlands have a significant effect on the risk management practices and derivatives use of US and Dutch firms.risk management;hedging;derivatives
Going Beyond Counting First Authors in Author Co-citation Analysis
Get PDFThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Neural ODE Processes
Get PDFNeural Ordinary Differential Equations (NODEs) use a neural network to model the instantaneous rate of change in the state of a system. However, despite their apparent suitability for dynamics-governed time-series, NODEs present a few disadvantages. First, they are unable to adapt to incoming data-points, a fundamental requirement for real-time applications imposed by the natural direction of time. Second, time-series are often composed of a sparse set of measurements that could be explained by many possible underlying dynamics. NODEs do not capture this uncertainty. In contrast, Neural Processes (NPs) are a new class of stochastic processes providing uncertainty estimation and fast data-adaptation, but lack an explicit treatment of the flow of time. To address these problems, we introduce Neural ODE Processes (NDPs), a new class of stochastic processes determined by a distribution over Neural ODEs. By maintaining an adaptive data-dependent distribution over the underlying ODE, we show that our model can successfully capture the dynamics of low-dimensional systems from just a few data-points. At the same time, we demonstrate that NDPs scale up to challenging high-dimensional time-series with unknown latent dynamics such as rotating MNIST digits
On second order behaviour in augmented neural ODEs
Get PDFNeural Ordinary Differential Equations (NODEs) are a new class of models that transform data continuously through infinite-depth architectures. The continuous nature of NODEs has made them particularly suitable for learning the dynamics of complex physical systems. While previous work has mostly been focused on first order ODEs, the dynamics of many systems, especially in classical physics, are governed by second order laws. In this work, we consider Second Order Neural ODEs (SONODEs). We show how the adjoint sensitivity method can be extended to SONODEs and prove that the optimisation of a first order coupled ODE is equivalent and computationally more efficient. Furthermore, we extend the theoretical understanding of the broader class of Augmented NODEs (ANODEs) by showing they can also learn higher order dynamics with a minimal number of augmented dimensions, but at the cost of interpretability. This indicates that the advantages of ANODEs go beyond the extra space offered by the augmented dimensions, as originally thought. Finally, we compare SONODEs and ANODEs on synthetic and real dynamical systems and demonstrate that the inductive biases of the former generally result in faster training and better performance
Inquiry into the interlocution of students engaged with mathematics: appreciating links between research and practice
No full textFor either to be useful, links between research and practice are critical. Just as important are connections between the practice of students engaged in mathematical activity and research that seeks to understand that practice. This research report explores lessons that researchers and practitioners can learn from an inquiry into the interlocution of students working collaboratively in small groups when engaged in talking and listening to each other. We use the term interlocution to denote discursive practices of learners in conversational exchanges. Questions that motivate this research included the following. What discursive practices do interlocutors employ as they work collaboratively to understand and resolve mathematical tasks? How do these practices influence the growth of their mathematical ideas? In what ways do their discursive practices help them move from a contextualized, situated task to generalize the task or their solution? Do students' discursive practices assist them to connect and generalize ideas from a new problem to others on which they have worked?Powell, A. B., & Maher, C. A. (2002). Inquiry into the interlocution of students engaged with mathematics: Appreciating links between research and practice. In D.S. Mewborn, P. Sztajn, D.Y. White, H.G. Wiegel, R.L. Bryant & K. Nooney (Eds.), Proceedings of the twenty-fourth annual meeting of the North American Chapter of the International Group for the Psychology of Mathematics Education (Athens, Georgia) (Vol. 1, pp. 317-329). Columbus, OH: ERIC Clearinghouse for Science, Mathematics, and Environmental Education
Dispelling the Myths Behind First-author Citation Counts
Get PDFWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
Interpretation of Recent Unrest Events (Bradyseism) at Campi Flegrei, Napoli (Italy): Comparison of Models Based on Cyclical Hydrothermal Events versus Shallow Magmatic Intrusive Events
Get PDFSeveral recent models that have been put forth to explain bradyseism at Campi Flegrei (CF), Italy, are discussed. Data obtained during long-term monitoring of the CF volcanic district has led to the development of a model based on lithological-structural and stratigraphic features that produce anisotropic and heterogeneous permeability features showing large variations both horizontally and vertically; these data are inconsistent with a model in which bradyseism is driven exclusively by shallow magmatic intrusions. CF bradyseism events are driven by cyclical magmatic-hydrothermal activity. Bradyseism events are characterized by cyclical, constant invariant signals repeating over time, such as area deformation along with a spatially well-defined seismogenic volume. These similarities have been defined as “bradyseism signatures” that allow us to relate the bradyseism with impending eruption precursors. Bradyseism is governed by an impermeable shallow layer (B-layer), which is the cap of an anticlinal geological structure culminating at Pozzuoli, where maximum uplift is recorded. This B-layer acts as a throttling valve between the upper aquifer and the deeper hydrothermal system that experiences short (1-102 yr) timescale fluctuations between lithostatic/hydrostatic pressure. The hydrothermal system also communicates episodically with a cooling and quasi-steady-state long timescale (103-104 yr) magmatic system enclosed by an impermeable carapace (A layer). Connectivity between hydrostatic and lithostatic reservoirs is episodically turned on and off causing alternatively subsidence (when the systems are connected) or uplift (when the systems are disconnected), depending on whether permeability by fractures is established or not. Earthquake swarms are the manifestation of hydrofracturing which allows fluid expansion; this same process promotes silica precipitation that seals cracks and serves to isolate the two reservoirs. Faults and fractures promote outgassing and reduce the vertical uplift rate depending on fluid pressure gradients and spatial and temporal variations in the permeability field. The miniuplift episodes also show “bradyseism signatures” and are well explained in the context of the short timescale process
Ground movement (bradysesim) in the Campi Flegrei volcanic area: a review. In: Vesuvius, Campi Flegrei, and Campanian volcanism.
No full textRe-equilibration of primary fluid inclusions in peritectic garnet from metapelitic enclaves, El Hoyazo, Spain
No full textPrimary-appearing fluid (Fl) and melt (FI) inclusions occur in peritectic garnet from restitic enclaves from El Hoyazo (Spain). The inclusions were trapped under conditions of immiscibility during partial melting of the enclaves. Trapped fluids in Bt-Grt-Sil and Spl-Crd enclaves have been characterized by microthermometric. Raman spectroscopic, electron microprobe (EMP) and transmission electron microprobe (TEM) analyses to better constrain melt and fluid products and pressure conditions of the partial melting event. In Bt-Grt-Sil enclaves, Fl are one phase and contain a CO(2)-N(2) mixture, sometimes with graphite as trapped phase. In Spl-Crd enclaves, Fl are two phase and contain an H(2)O-rich (<= 90 mol%), with minor amounts of CO(2), N(2), and traces of H(2)S and CH(4). Graphite often occurs as a trapped phase in the H(2)O-rich Fl, and rare carbonates and other accessory minerals are also observed. Although decrepitation features are not recognized during examination with a petrographic microscope, Fl densities based on mass balance constraints are always lower than expected at the inferred PT conditions of entrapment, 5-7 kbar and 800-900 degrees C. Extremely low densities (approximate to 0.1 g cm(-1)) of Fl in Bt-Grt-Sil enclaves suggest a pressure <= 500 bar at 800-900 degrees C, while densities up to 0.53 g cm(-1) in Spl-Crd enclaves indicate P kbar at 800-900 degrees C. Re-equilibration is likely to have occurred via partial decrepitation, as suggested by TEM studies that show rare partially annealed sub-pm cracks, containing small cavities, which may have been the pathways for fluid movement out of the inclusions. MI coexisting with Fl have a rhyolitic, peraluminous composition, with higher H(2)O contents of MI in Spl-Crd enclaves (approximate to 9 wt.%) compared to MI in Bt-Grt-Sil enclaves (approximate to 3 wt.%). Based on published data, peritectic garnet in Spl-Crd enclaves grew in the presence of a leucogranitic melt saturated in an H(2)O-rich fluid, in good agreement with the inferred garnet PT growth conditions. The composition of the fluid phase coexisting with melt in Bt-Grt-Sil enclaves cannot be evaluated owing to the almost complete decrepitation and fluid loss from Fl, and may only be inferred to have been more CO(2)-rich, based on the lower H(2)O content of the coexisting melt
Proximal Distilled Evolutionary Reinforcement Learning
No full textReinforcement Learning (RL) has achieved impressive performance in many complex environments due to the integration with Deep Neural Networks (DNNs). At the same time, Genetic Algorithms (GAs), often seen as a competing approach to RL, had limited success in scaling up to the DNNs required to solve challenging tasks. Contrary to this dichotomic view, in the physical world, evolution and learning are complementary processes that continuously interact. The recently proposed Evolutionary Reinforcement Learning (ERL) framework has demonstrated mutual benefits to performance when combining the two methods. However, ERL has not fully addressed the scalability problem of GAs. In this paper, we show that this problem is rooted in an unfortunate combination of a simple genetic encoding for DNNs and the use of traditional biologically-inspired variation operators. When applied to these encodings, the standard operators are destructive and cause catastrophic forgetting of the traits the networks acquired. We propose a novel algorithm called Proximal Distilled Evolutionary Reinforcement Learning (PDERL) that is characterised by a hierarchical integration between evolution and learning. The main innovation of PDERL is the use of learning-based variation operators that compensate for the simplicity of the genetic representation. Unlike traditional operators, our proposals meet the functional requirements of variation operators when applied on directly-encoded DNNs. We evaluate PDERL in five robot locomotion settings from the OpenAI gym. Our method outperforms ERL, as well as two state-of-the-art RL algorithms, PPO and TD3, in all tested environments
- …
