Non-Parametric Regression Methods

The nature of the financial time series is complex, continuous interchange of stochastic and deterministic regimes. Therefore, it is difficult to forecast with parametric techniques. Instead of parametric models, we propose three techniques and compare with each other. Neural networks and support vector regression (SVR) are two universally approximators. They are data-driven non parametric models. ARCH/GARCH models are also investigated. Our assumption is that the future value of Istanbul Stock Exchange 100 index daily return depends on the financial indicators although there is no known parametric model to explain this relationship. This relationship comes from the technical analysis. Comparison shows that the multi layer perceptron networks overperform the SVR and time series model (GARCH).     

Distress among disaster-affected populations: delay in relief provision

Central to humanitarian logistics is the minimization of distress among impacted populations in the aftermath of a disaster. In this paper, we characterize two levels of distress, termed criticality and destitution, with respect to the delay provision of relief items. Delay in provision of a relief item will lead to destitution for a tolerable number of days, beyond which it will lead to criticality. We develop a mixed-integer goal program that quantifies these two metrics with respect to the number of days without provision of each of a set of relief items. The model determines the allocation of resources and the distribution of available relief items in a manner that minimizes criticality and destitution in affected population segments. The use of the model is demonstrated for the aftermath of a catastrophic earthquake in Istanbul, expected to occur by 2030.     

Peripherally and Axially Carboxylic Acid Substituted Subphthalocyanines for Dye‐Sensitized Solar Cells

A series of subphthalocyanines (SubPcs) bearing a carboxylic acid group either at the peripheral or axial position have been designed and synthesized to investigate the influence of the COOH group positions on the dye‐sensitized solar cell (DSSC) performance. The DSSC devices based on SubPcs with axially substituted carboxylic acid groups showed low photovoltaic performance, whereas peripherally substituted one exhibited higher power conversion efficiency owing to improved injection from LUMO of SubPcs to the TiO₂ conduction band.     

Excitation pulse deconvolution in luminescence lifetime analysis for oxygen measurements in vivo

Oxygen-dependent quenching of phosphorescence has been proven to be a valuable tool for the measurement of oxygen concentrations both in vitro and in vivo. For biological measurements the relatively long lifetimes of phosphorescence have promoted time-domain-based devices using xenon arc flashlamps as the most common excitation light source. The resulting complex form of the excitation pulse leads to complications in the analysis of phosphorescence lifetimes and ultimately to errors in the recovered pO2 values. Although the problem has been recognized, the consequences on in vivo phosphorescence lifetime measurements have been neglected so far. In this study, the consequences of finite excitation flash duration are analyzed using computer simulations, and a method for the recovery of phosphorescence decay times from complex photometric signals is presented. The analysis provides an explanation as to why different calibration constants are reported in the literature and presents a unified explanation whereby calibration constants are not solely a property of the dye but also of the measuring device. It is concluded that complex excitation pulse patterns without appropriate analysis methods lead to device-specific calibration constants and nonlinearity and can be a potent source of errors when applied in vivo. The method of analysis presented in this article allows reliable phosphorescence lifetime measurements to be made for oxygen pressure measurements and can easily be applied to existing phosphorimeters.     

Impacts of pH and molecular structure on ultrasonic degradation of azo dyes

Sonochemical bleaching of monoazo dyes was investigated by irradiating 30 microM solutions of two "aryl-azo-naphthol" type model dyes in acidic, neutral and basic conditions using a 300 kHz emitter. It was found that the rate of bleaching in all cases was first order with respect to the maximum absorption of the dye in the visible band, and accelerated with increased acidity. The inhibition observed at alkaline conditions was attributed to the formation of anionic dye structures and their competition with the dye and its intermediate oxidation products for hydroxyl radicals, which are the major precursors of azo dye oxidation in sonicated water. Decolorization of the dyes was also related to the size of the molecule and the type or position of substituents about azo bonds. Comparison of color decay rates at similar conditions showed that the dye with a simple structure, low molecular mass and one ortho-substituent (hydroxyl) about the azo bond bleached considerably faster than the one having a more complicated structure (higher mass) and an additional o-substituent to the azo bond other than the OH group.     

A numerical method for determining refractive index of a glass sample from its implicit transcendental function

The refractive index of a glass sample was determined from an implicit function of its optical path within the sample arm of a Michelson interferometer.On rotation of the sample from normal incidence, the light beam suffers increasing refraction, causing the optical path for air to decrease whilst that for the glass sample increases. This is observed as a shift in the number of fringes, which were captured and counted in real time on a computer, as rotation proceeded. The angle of rotation and the fringe shift were entered into an implicit function of optical path versus refractive index written to an Excel worksheet. A refractive index matching the wavelength of the He-Ne laser light source was read-off to three decimal places. A new source of uncertainty has been identified, misalignment on a micrometer scale of the laser normal to the sample surface. Whilst driving the sample to find normal incidence a finite angular region of insensitivity (dead space) occurs due to misalignment of the laser normal to the sample surface by a few micrometers. A linearization technique was employed to compensate for this offset and hence reduce its uncertainty contribution.     

Role of the n+1 amino acid residue on the deamidation of asparagine in pentapeptides

Deamidation plays an important role in biochemical phenomena such as aging. The role of the n + 1 residue on the deamidation of asparagine (asparagine being the nth residue) in three pentapeptide chains (GGNGG, GGNMG and GGNIG) has been analysed with hybrid computational tools. Potentials of mean force at 300 K were calculated from the MD/replica exchange simulations using weighted histogram analysis (WHAM) in explicit water. The snapshots were clustered taking into account the requirements of the plausible deamidation mechanisms, as such the tautomerisation of the asparagine side chain as initial step has been confirmed, based on the proximity of water to the deamidation site. The ultimate goal being to gain an insight on the peptide backbone N-H acidity, quantum mechanical calculations have been carried out. For this purpose, the distribution of Φ/Ψ, Φ₂/Ψ and end-to-end distances deduced from the WHAM diagrams have been considered and a total of 110 structures have been sampled. These neutral pentapeptides as well as their corresponding anions have been optimised (B3LYP/6-31++G(d,p)) in implicit water in order to gain an insight on the peptide backbone N-H acidity. In this study, we have shown that the open conformations of the neutrals and the anions, which display a β sheet like structure are well populated and their pK_as rank in the same order as the deamidating half-lives, that is the peptides that deaminate fastest can more readily access conformations that are more acidic.     

Phthalocyanines: colorful macroheterocyclic sensitizers for dye-sensitized solar cells

Abstract  

Dye-sensitized solar cells (DSSCs) have gained great attention as lower-cost alternatives to conventional photovoltaic devices. One way to improve the excellent efficiencies (ca. 11%) exhibited by DSSCs based on ruthenium polypyridyl dyes would consist in using sensitizers with enhanced light-harvesting properties in the red region of the spectrum. Phthalocyanines (Pcs) are very robust molecules which present extraordinary high extinction coefficients in the 600- to 700-nm spectral region. Intensive research has been focused on reducing the undesired aggregation phenomena of Pcs on the metal oxide surface, while keeping a good electronic coupling between the LUMO of the Pcs and the TiO₂ conduction band, and a good solubility of the dye in organic solvents. Recently, unsymmetrically substituted “push–pull” Pcs have emerged as efficient red-absorbing dyes, reaching power conversion efficiencies of up to 4.7%, when used as single sensitizers.