The spin-

In recent years, financial regulations such as Basel II and Solvency II have highlighted the utility of credit risk assessments through internal rating systems, particularly for estimating the probability of default (PD) of credit exposures.     

Modelling of insurers’ rating determinants. An application of machine learning techniques and statistical models

In recent years, financial regulations such as Basel II and Solvency II have highlighted the utility of credit risk assessments through internal rating systems, particularly for estimating the probability of default (PD) of credit exposures.     

Statistical models of diffusion and aggregation for coke formation in a catalyst pore

We simulated models of diffusion and aggregation in long pores of small widths in order to represent the basic mechanisms of coke deposition in catalysts’ pores. Coke precursors are represented by particles injected at the pore entrance. Knudsen diffusion, which is usually expected inside the pores, is modeled by ballistic motion of those particles. The regime of molecular diffusion is also analyzed via models of lattice random walks biased along the pores. The aggregation at the surface or near previously aggregated particles was modeled by different probabilistic rules, accounting for the possibilities of more compact or more ramified deposits. In the model of Knudsen diffusion and in some cases of molecular diffusion, there is an initial regime of uniform deposition along the pore, after which the deposits acquire an approximately wedge shape, with the pore plugging near its entrance. After the regime of uniform deposition and before that of critical pore plugging, the average aggregation position slowly decreases with the number N of deposited particles approximately as N^-0.25. The apparently universal features of deposits generated by microscopic models are compared with those currently adopted in continuum models.     

On using non-Gaussian distributions to perform statistical linearization

In statistical linearization non-linear elements are approximated by equivalent linear elements according to recipes proposed by the pioneers of the procedure. The recipes require the evaluation of certain statistics which, ideally, should be evaluated using the exact probability distribution of the non-linear response. Because the exact non-linear response distribution is unknown it has become traditional to use a Gaussian distribution as an approximation to the exact distribution. With the modern computing tools now available it is easy to use non-Gaussian distributions which can provide better approximations in cases where Gaussian distributions are not appropriate. Examples are displayed for power-law oscillators with stiffening and softening springs, and for the Duffing oscillator, and for a double-well oscillator. Two families of probability distributions with varying shape are studied.     

Prediction of work-related accidents according to working conditions using support vector machines

Support vector machines (SVMs), which are a kind of statistical learning methods, were applied in this research work to predict occupational accidents with success. In the first place, semi-parametric principal component analysis (SPPCA) was used in order to perform a dimensional reduction, but no satisfactory results were obtained. Next, a dimensional reduction was carried out using an innovative and intelligent computing regression algorithm known as multivariate adaptive regression splines (MARS) model with good results. The variables selected as important by the previous MARS model were taken as input variables for a SVM model. This SVM technique was able to classify, according to their working conditions, those workers that have suffered a work-related accident in the last 12 months and those that have not. SVM technique does not over-fit the experimental data and gives place to a better performance than back-propagation neural network models. Finally, the results and conclusions of this study are presented.     

A decision theory for partially consonant belief functions

Partially consonant belief functions (pcb), studied by Walley, are the only class of Dempster-Shafer belief functions that are consistent with the likelihood principle of statistics. Structurally, the set of foci of a pcb is partitioned into non-overlapping groups and within each group, foci are nested. The pcb class includes both probability function and Zadeh’s possibility function as special cases. This paper studies decision making under uncertainty described by pcb. We prove a representation theorem for preference relation over pcb lotteries to satisfy an axiomatic system that is similar in spirit to von Neumann and Morgenstern’s axioms of the linear utility theory. The closed-form expression of utility of a pcb lottery is a combination of linear utility for probabilistic lottery and two-component (binary) utility for possibilistic lottery. In our model, the uncertainty information, risk attitude and ambiguity attitude are separately represented. A tractable technique to extract ambiguity attitude from a decision maker behavior is also discussed.     

Calculating thermodynamic properties from perturbation theory: II. An analytic representation for the square-well chain fluid

An analytic representation of thermodynamic properties of the freely jointed square-well chain fluid is developed based on the thermodynamic perturbation theory of Barker–Henderson, Zhang and Weitheim. By using a real function expression for the radial distribution function and incorporating structural information for square-well monomer of TPT1 model, an analytic expression for the Helmholtz energy of square-well chain fluid is expanded from Zhang’s analytic expressions for thermodynamic properties of square-well monomer. The expression leads to good predictions of the compressibility factor, residual internal energy and constant-volume heat capacity for 4-mer, 8-mer and 16-mer square-well fluids when compared with the Monte Carlo (MC) simulation results. The incorporating structural information for square-well dimer of TPT-D model is also calculated. To obtain the constant-volume heat capacity needed, NVT MC simulations were performed.     

The Order Parameter of the Chiral Potts Model

An outstanding problem in statistical mechanics is the order parameter of the chiral Potts model. An elegant conjecture for this was made in 1983. It has since been successfully tested against series expansions, but there is as yet no proof of the conjecture. Here we show that if one makes a certain analyticity assumption similar to that used to derive the free energy, then one can indeed verify the conjecture. The method is based on the ‘‘broken rapidity line’’ approach pioneered by Jimbo et al. (J. Phys. A 26:2199--2210 (1993).).     

The third industrial fluid properties simulation challenge

The third industrial fluid properties simulation challenge was held from March to September 2006. As in the previous two events contestants were challenged to predict specific, industrially relevant, properties of fluid systems. Their efforts were judged based on the agreement of the predicted values with previously unpublished experimental data (provided by researchers at ExxonMobil and DuPont). The focus of this contest was on the transferability of modeling methods—the ability to predict properties for materials that are chemically different, or at different state points, to those used in model parameterization and validation. Nine groups attempted to compute bubble point pressures for mixtures of 1,1,1,2,3,3,3-heptafluoropropane (HFC-227ea) and ethanol at 343 K, given data for mixtures at 283 K, and given the pure component vapor pressures. They used a range of different techniques including statistical mechanical and molecular simulations-based approaches. Four of the groups were recognized for providing predictions that were significantly more accurate than would be obtained by extrapolation using the NRTL model (the standard engineering approach). Three groups undertook the more challenging “molecular transferability” problem, attempting to predict shear viscosities at two different state points for a range of diols and triols for which little experimental data was available.     

Development of the Monte Carlo model CASCADE-2004 of high-energy nuclear interactions

The Monte Carlo code CASCADE for the calculation of inelastic hadron- and nucleus-nucleus interactions at energies from several tens of MeV up to several tens of GeV and for modelling of nuclear-physical processes accompanying the transport of particles and nuclei in matter is improved by considering a more detailed model of decay of highly excited residual (after-intranuclear-cascade) nuclei. Results of calculations are in good agreement with experiment. However, there are some deviations for light-isotope production, which prompt the necessity of developing more correct models of evaporation and strong asymmetric high-energy fission.