: This thesis applies neural network feature selection techniques to multivariate time series data to improve prediction of a target time series. Two approaches to feature selection are used. First, a subset enumeration method is used to determine which financial indicators are most useful for aiding in prediction of the S&P 500 futures daily price. The candidate indicators evaluated include RSI, Stochastics and several moving averages. Results indicate that the Stochastics and RSI indicators result in better prediction results than the moving averages. The second approach to feature selection is calculation of individual saliency metrics. A new decision boundary-based individual saliency metric, and a classifier independent saliency metric are developed and tested. Ruck's saliency metric, the decision boundary based saliency metric, and the classifier independent saliency metric are compared for a data set consisting of the RSI and Stochastics indicators as well as delayed closing price values. The decision based metric and the Ruck metric results are similar, but the classifier independent metric agrees with neither of the other metrics. The nine most salient features, determined by the decision boundary based metric, are used to train a neural network and the results are presented and compared to other published results. (AN)

/pdf/nonlinear-time-series-analysis-1j3a35n4y3.pdf

Nonlinear Time Series Analysis.

Amphiphiles are synthetic or natural molecules with the ability to self-assemble into a wide variety of structures including micelles, vesicles, nanotubes, nanofibers, and lamellae. Self-assembly processes of amphiphiles have been widely used to mimic biological systems, such as assembly of lipids and proteins, while their integrated actions allow the performance of highly specific cellular functions which has paved a way for bottom-up bionanotechnology. While amphiphiles self-assembly has attracted considerable attention for decades due to their extensive applications in material science, drug and gene delivery, recent developments in nanoscience stimulated the combination of the simple approaches of amphiphile assembly with the advanced concept of supramolecular self-assembly for the development of more complex, hierarchical nanostructures. Introduction of stimulus responsive supramolecular amphiphile assembly-disassembly processes provides particularly novel approaches for impacting bionanotechnology applications. Leading examples of these novel self-assembly processes can be found, in fact, in biosystems where assemblies of different amphiphilic macrocomponents and their integrated actions allow the performance of highly specific biological functions. In this perspective, we summarize in this tutorial review the basic concept and recent research on self-assembly of traditional amphiphilic molecules (such as surfactants, amphiphile-like polymers, or lipids) and more recent concepts of supramolecular amphiphiles assembly which have become increasingly important in emerging nanotechnology.

/pdf/amphiphiles-self-assembly-basic-concepts-and-future-bl2hcjeioo.pdf

Amphiphiles Self-Assembly: Basic Concepts and Future Perspectives of Supramolecular Approaches

Ionic liquids confined in porous matrices: Physicochemical properties and applications

The production of strings (disclination lines and loops) has been observed by means of the Kibble mechanism of domain (bubble) formation in the isotropic-nematic phase transition of the uniaxial nematic liquid crystal 4-cyano-4'-n-pentylbiphenyl. The number of strings formed per bubble is about 0.6. This value is in reasonable agreement with a numerical simulation of the experiment in which the Kibble mechanism is used for the order parameter space of a uniaxial nematic liquid crystal.

The Cosmological Kibble Mechanism in the Laboratory: String Formation in Liquid Crystals

Studies have repeatedly reported that math and science are perceived as male domains, and scientists as predominantly male. However, the impact of the gender image of school science subjects on young people’s career choice has not yet been analyzed. This paper investigates the impact of the masculinity image of three school subjects – chemistry, mathematics and physics – on secondary students’ career aspirations in STEM fields. The data originated from a cross-sectional study among 1’364 Swiss secondary school students who were close to obtaining their matriculation diploma. By means of a standardized survey, data on students’ perception of masculinity of science school subjects were collected using semantic differentials. The results indicate that for both sexes, math has the strongest masculinity attribution, followed by physics as second, and, finally, chemistry with the lowest masculinity attribution. With respect to gender differences, our findings have shown that among female students, the attribution of masculinity to the three school subjects does not differ significantly, meaning that female students rated all subjects similarly strongly as masculine. Within the group of male students however, the attribution of masculinity to math compared to chemistry and physics differs significantly, whereas the attribution of masculinity to chemistry and physics does not. Our findings also suggest that gender-science stereotypes of math and science can potentially influence young women’s and men’s aspirations to enroll in a STEM major at university by showing that a less pronounced masculine image of science has the potential to increase the likelihood of STEM career aspirations. Finally, the paper discusses ways of changing the image of math and science in the context of secondary education in order to overcome the disparities between females and males in STEM.

/pdf/the-gender-gap-in-stem-fields-the-impact-of-the-gender-3549exajjd.pdf

The Gender Gap in STEM Fields: The Impact of the Gender Stereotype of Math and Science on Secondary Students' Career Aspirations

Finance and economics are complex nonlinear systems that are affected by various external factors, including of course human action, bilateral relations, conflicts, and policy. Time delays in a financial system take into account the amount of time that passes from a particular policy or decision being made to it actually taking effect. It is thus important to consider time delays as an integral part of modeling in this field. Moreover, many features of financial systems cannot be expressed sufficiently precisely by means of integer-order calculus. Fractional-order calculus alleviates these shortcomings. The aim of this paper is therefore to study the dynamics and complexity in a fractional-order financial system with time delays. We observe fascinating transitions to deterministic chaos, including cascading period doubling, as well as high levels of complexity. This is particularly true in response to variations of derivative orders, which are thus identified as key system parameters.

Chaos and complexity in a fractional-order financial system with time delays

Synchronizability of two neurons with switching in the coupling

Liquid-crystal phases are typical examples of soft and complex materials that exhibit an abundance of different phenomena. In this paper we present some of our results contributing to the understanding of the physics of defects in nematic liquid crystals. The examples presented exhibit many features that are also of interest for other branches of physics. We describe nematic point defects, the annihilation dynamics of a defect and anti-defect pair, and the coarsening dynamics of a dense pattern of defects after a sudden symmetry-breaking phase transition.

Physics of defects in nematic liquid crystals

Bogdanov–Takens singularity in the Hindmarsh–Rose neuron with time delay

Over the past few decades, many works have been devoted to designing simple chaotic systems based on analog electronic circuits. However, the same attention is not observed in digital chaotic systems. This paper presents a design of a digital chaotic system using a digit complement. This special case of fixed-point number representation allows us to reduce the silicon area and the number of logic elements to perform the arithmetic operations. The design presents a configurable number of bits, and it is based on the logistic map. The proposed circuit has been implemented on a reconfigurable hardware, FPGA Cyclone V, showing that the number of logic elements has been significantly reduced compared to other works in the literature.

Robert Repnik

Papers

Chaos and complexity in a fractional-order financial system with time delays

Synchronizability of two neurons with switching in the coupling

Physics of defects in nematic liquid crystals

Bogdanov–Takens singularity in the Hindmarsh–Rose neuron with time delay

Minimal digital chaotic system