Large Set Of Models Research Articles

Manipulating and measuring variation in deep neural network (DNN) representations of objects

We explore how DNNs can be used to develop a computational understanding of individual differences in high-level visual cognition given their ability to generate rich meaningful object representations informed by their architecture, experience, and training protocols. As a first step to quantifying individual differences in DNN representations, we systematically explored the robustness of a variety of representational similarity measures: Representational Similarity Analysis (RSA), Centered Kernel Alignment (CKA), and Projection-Weighted Canonical Correlation Analysis (PWCCA), with an eye to how these measures are used in cognitive science, cognitive neuroscience, and vision science. To manipulate object representations, we next created a large set of models varying in random initial weights and random training image order, training image frequencies, training category frequencies, and model size and architecture and measured the representational variation caused by each manipulation. We examined both small (All-CNN-C) and commonly-used large (VGG and ResNet) DNN architectures. To provide a comparison for the magnitude of representational differences, we established a baseline based on the representational variation caused by image-augmentation techniques used to train those DNNs. We found that variation in model randomization and model size never exceeded baseline. By contrast, differences in training image frequency and training category frequencies caused representational variation that exceeded baseline, with training category frequency manipulations exceeding baseline earlier in the networks. These findings provide insights into the magnitude of representational variations that can be expected with a range of manipulations and provide a springboard for further exploration of systematic model variations aimed at modeling individual differences in high-level visual cognition.

Z’ boson mass reach and model discrimination at muon colliders

We study the discrimination power of future multi-TeV muon colliders for a large set of models with extended gauge symmetries and additional neutral gauge bosons (“Z′\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$Z'$$\\end{document}-models”). Our study is carried out using a χ2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\chi ^2$$\\end{document}-analysis of leptonic observables of s-channel scattering in effective Z′\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$Z'$$\\end{document}-models. We make use of angular and chiral asymmetries induced in such models to find the discovery reach of a given muon collider setup in terms of the Z′\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$Z'$$\\end{document} mass and to discriminate between the different scenarios. In this context, we discuss how polarized beams – should they become available at muon colliders – or polarization measurements can help in the discrimination. Our results show that typical muon collider setups which are currently under consideration can give a significantly higher reach compared to existing bounds and projections for high-luminosity LHC.

Modeling prevention behaviors during the COVID-19 pandemic using Bayesian belief networks and protection motivation theory.

Prevention behaviors are important in mitigating the transmission of COVID-19. The protection motivation theory (PMT) links perceptions of risk and coping ability with the act of adopting prevention behaviors. The goal of this research is to test the application of the PMT in predicting adoption of prevention behaviors during the COVID-19 pandemic. Two research objectives are achieved to explore motivating factors for adopting prevention behaviors. (1) The first objective is to identify variables that are strong predictors of prevention behavior adoption. A data-driven approach is used to train Bayesian belief network (BBN) models using results of a survey of participants reporting risk perceptions and prevention behaviors during the COVID-19 pandemic. A large set of models are generated and analyzed to identify significant variables. (2) The second objective is to develop models based on the PMT to predict prevention behaviors. BBN models that predict prevention behaviors were developed using two approaches. In the first approach, a data-driven methodology trains models using survey data alone. In the second approach, expert knowledge is used to develop the structure of the BBN using PMT constructs. Results demonstrate that trust and experience with COVID-19 were important predictors for prevention measure adoption. Models that were developed using the PMT confirm relationships between coping appraisal, threat appraisal, and protective behaviors. Data-driven and PMT-based models perform similarly well, confirming the use of PMT in this context. Predicting adoption of social distancing behaviors provides insight for developing policies duringpandemics.

Search-Based Software Testing Driven by Automatically Generated and Manually Defined Fitness Functions

Search-based software testing (SBST) typically relies on fitness functions to guide the search exploration toward software failures. There are two main techniques to define fitness functions: (a) automated fitness function computation from the specification of the system requirements, and (b) manual fitness function design. Both techniques have advantages. The former uses information from the system requirements to guide the search toward portions of the input domain more likely to contain failures. The latter uses the engineers’ domain knowledge. We propose ATheNA , a novel SBST framework that combines fitness functions automatically generated from requirements specifications and those manually defined by engineers. We design and implement ATheNA-S , an instance of ATheNA that targets Simulink ® models. We evaluate ATheNA-S by considering a large set of models from different domains. Our results show that ATheNA-S generates more failure-revealing test cases than existing baseline tools and that the difference between the runtime performance of ATheNA-S and the baseline tools is not statistically significant. We also assess whether ATheNA-S could generate failure-revealing test cases when applied to two representative case studies: one from the automotive domain and one from the medical domain. Our results show that ATheNA-S successfully revealed a requirement violation in our case studies.

On the core-collapse supernova explanation for LAMOST J1010 + 2358

ABSTRACT Low-metallicity very massive stars with an initial mass of ∼140–$260\, \mathrm{M}_\odot$ are expected to end their lives as pair-instability supernovae (PISNe). The abundance pattern resulting from a PISN differs drastically from regular core-collapse supernova (CCSN) models and is expected to be seen in very metal-poor (VMP) stars of [Fe/H] ≲ −2. Despite the routine discovery of many VMP stars, the unique abundance pattern expected from PISNe has not been unambiguously detected. The recently discovered VMP star LAMOST J1010 + 2358, however, shows a peculiar abundance pattern that is remarkably well fit by a PISN, indicating the potential first discovery of a bonafide star born from gas polluted by a PISN. In this paper, we study the detailed nucleosynthesis in a large set of models of CCSN of Pop III and Pop II star of metallicity [Fe/H] = −3 with masses ranging from 12 to $30\, \mathrm{M}_\odot$. We find that the observed abundance pattern in LAMOST J1010 + 2358 can be fit at least equally well by CCSN models of ∼12–$14\, \mathrm{M}_\odot$ that undergo negligible fallback following the explosion. The best-fitting CCSN models provide a fit that is even marginally better than the best-fitting PISN model. We conclude the measured abundance pattern in LAMOST J1010 + 2358 could have originated from a CCSN and therefore cannot be unambiguously identified with a PISN given the set of elements measured in it to date. We identify key elements that need to be measured in future detections in stars like LAMOST J1010 + 2358 that can differentiate between CCSN and PISN origin.

Is distance from equilibrium a good indicator for a reaction’s flux control?

By analysing a large set of models obtained from the JWS Online and Biomodels databases, we tested to what extent the disequilibrium ratio can be used as an estimator for the flux control of a reaction, a discussion point that was already raised by Kacser and Burns, and Heinrich and Rapoport in their seminal MCA manuscripts. Whereas no functional relation was observed, the disequilibrium ratio can be used as an estimator for the maximal flux control of a reaction step. We extended the original analysis of the relationship by incorporating the overall pathway disequilibrium ratio in the expression, which made it possible to make explicit expressions for flux control coefficients.

Tracking of Maneuvering Targets Using a Variable Structure Multiple Model Algorithm

Introduction. In recent years, much attention has been paid to the development of trajectory filtering methods for tracking maneuvering targets. Multi-model (MM) algorithms are widely used for filtering maneuvering targets. Conventional MM algorithms are characterized by a fixed structure. However, highly maneuvering targets require a sufficiently large set of models covering the entire range of possible maneuvers, although an increase in the number of models cannot ensure an increase in the accuracy of tracking. To overcome these problems, multiple model algorithms with a variable structure (VSMM) were proposed. This article proposes two VSMM algorithms for tracking maritime targets performing a coordinated turn at constant speed. These are algorithms with a variable set of models based on adaptive grid and switching grid methods.Aim. To develop an adaptive trajectory tracking algorithm that uses a constant turn model to track maneuvering surface objects.Materials and methods. The resulting algorithm is based on the theory of grid adaptation in multi-model estimation methods and is used to estimate the components of the coordinate and velocity vectors of surface maneuvering targets. The algorithm efficiency was evaluated using computer statistical modeling in the MATLAB environment.Results. The structure of an adaptive VSMM algorithm was described. Simulations were carried out to confirm the algorithm efficiency. In the considered simulation scenarios, the algorithm produces effective estimates of the coordinate vectors and speed of surface maneuvering targets.Conclusion. Adaptive algorithms improve the efficiency of target tracking in comparison with multi-model algorithms with a fixed structure, at the same time as saving computational resources.

Top partners and scalar dark matter: A nonminimal reappraisal

A large set of models beyond the Standard Model of particle physics suggest that the top quark plays a special role in fundamental interactions. At the same time some of these models predict that a particle responsible for dark matter is present in the physical spectrum of the theory, and in particular cases dark matter may be specially linked to the top sector. Such a new physics configuration occurs, for instance, in composite frameworks that additionally feature vector-like quarks as mediators of the interactions of the dark matter with the Standard Model. We investigate the interplay of top-philic dark matter and the presence of several vector-like top partners. Such a non-minimal setup with several vector-like fermions is typical of composite scenarios, and we make use of an effective model to study to which extent such non-minimality modifies the collider and cosmological phenomenology of the minimal model, the latter only including one vector-like mediator of the dark matter interactions with the Standard Model.

Pulse profile modelling of thermonuclear burst oscillations − I. The effect of neglecting variability

ABSTRACT We study the effects of the time-variable properties of thermonuclear X-ray bursts on modelling their millisecond-period burst oscillations. We apply the pulse profile modelling technique that is being used in the analysis of rotation-powered millisecond pulsars by the Neutron Star Interior Composition Explorer to infer masses, radii, and geometric parameters of neutron stars. By simulating and analysing a large set of models, we show that overlooking burst time-scale variability in temperatures and sizes of the hot emitting regions can result in substantial bias in the inferred mass and radius. To adequately infer neutron star properties, it is essential to develop a model for the time-variable properties or invest a substantial amount of computational time in segmenting the data into non-varying pieces. We discuss prospects for constraints from proposed future X-ray telescopes.

Models of plate tectonics with the Lattice Boltzmann Method

Modern geodynamics is based on the study of a large set of models, with the variation of many parameters, whose analysis in the future will require Machine Learning to be analyzed. We introduce here for the first time how a formulation of the Lattice Boltzmann Method capable of modeling plate tectonics, with the introduction of plastic non-linear rheology, is able to reproduce the breaking of the upper boundary layer of the convecting mantle in plates. Numerical simulation of the earth’s mantle and lithospheric plates is a challenging task for traditional methods of numerical solution to partial differential equations (PDE’s) due to the need to model sharp and large viscosity contrasts, temperature dependent viscosity and highly nonlinear rheologies. Nonlinear rheologies such as plastic or dislocation creep are important in giving mantle convection a past history. We present a thermal Lattice Boltzmann Method (LBM) as an alternative to PDE-based solutions for simulating time-dependent mantle dynamics, and demonstrate that the LBM is capable of modeling an extremely nonlinear plastic rheology. This nonlinear rheology leads to the emergence plate tectonic like behavior and history from a two layer viscosity model. These results demonstrate that the LBM offers a means to study the effect of highly nonlinear rheologies on earth and exoplanet dynamics and evolution.

Accelerating Results in Carbon Storage Studies Using an Integrated and Automated Approach

Geological storage of carbon dioxide (CO2) is considered an enabler of different business models aligned with decarbonization of the energy market. Partnerships are forming worldwide to develop large-scale carbon capture, utilization and storage (CCUS) projects: 2021 was a record year for project pipeline growth for these types of projects. This growth will result in an increasing need for subsurface technologies that can unlock fast time-to-results throughout all the steps of the project, from site selection to storage monitoring. At an early stage of a carbon storage project, a thorough verification of the technical and economic viability of the project is critical. The high degree of geological uncertainties in the case of storage in under-explored saline aquifers can make this step challenging. As the project progresses, fast assimilation of monitoring data to prove conformance and update predictions of the storage complex performance is key. An advanced technology from AspenTech can serve as a catalyst for efficient carbon storage studies. It tightly integrates static and dynamic domains and offers the propagation of uncertainties, from seismic characterization through to geological modelling and simulation. Using results from a large set of models increases predictability of the subsurface and enables more efficient analysis of uncertainty in predicted storage capacity and containment. This fully automated workflow can be run at will with new data, drastically reducing the time needed by carbon storage teams to update the model and the predictions as monitoring data is acquired.

The Occurrence Birth-Death Process for Combined-Evidence Analysis in Macroevolution and Epidemiology.

Phylodynamic models generally aim at jointly inferring phylogenetic relationships, model parameters, and more recently, the number of lineages through time, based on molecular sequence data. In the fields of epidemiology and macroevolution, these models can be used to estimate, respectively, the past number of infected individuals (prevalence) or the past number of species (paleodiversity) through time. Recent years have seen the development of “total-evidence” analyses, which combine molecular and morphological data from extant and past sampled individuals in a unified Bayesian inference framework. Even sampled individuals characterized only by their sampling time, that is, lacking morphological and molecular data, which we call occurrences, provide invaluable information to estimate the past number of lineages. Here, we present new methodological developments around the fossilized birth–death process enabling us to (i) incorporate occurrence data in the likelihood function; (ii) consider piecewise-constant birth, death, and sampling rates; and (iii) estimate the past number of lineages, with or without knowledge of the underlying tree. We implement our method in the RevBayes software environment, enabling its use along with a large set of models of molecular and morphological evolution, and validate the inference workflow using simulations under a wide range of conditions. We finally illustrate our new implementation using two empirical data sets stemming from the fields of epidemiology and macroevolution. In epidemiology, we infer the prevalence of the coronavirus disease 2019 outbreak on the Diamond Princess ship, by taking into account jointly the case count record (occurrences) along with viral sequences for a fraction of infected individuals. In macroevolution, we infer the diversity trajectory of cetaceans using molecular and morphological data from extant taxa, morphological data from fossils, as well as numerous fossil occurrences. The joint modeling of occurrences and trees holds the promise to further bridge the gap between traditional epidemiology and pathogen genomics, as well as paleontology and molecular phylogenetics. [Birth–death model; epidemiology; fossils; macroevolution; occurrences; phylogenetics; skyline.]

Boosting the transferability of adversarial examples via stochastic serial attack

Deep neural networks (DNNs) are vulnerable to adversarial examples, which are crafted by imposing mild perturbation on clean ones. An intriguing property of adversarial examples is that they are efficient among different DNNs. Thus transfer-based attacks against DNNs become an increasing concern. In this scenario, attackers devise adversarial instances based on the local model without feedback information from the target one. Unfortunately, most existing transfer-based attack methods only employ a single local model to generate adversarial examples. It results in poor transferability because of overfitting to the local model. Although several ensemble attacks have been proposed, the transferability of adversarial examples merely have a slight increase. Meanwhile, these methods need high memory cost during the training process. To this end, we propose a novel attack strategy called stochastic serial attack (SSA). It adopts a serial strategy to attack local models, which reduces memory consumption compared to parallel attacks. Moreover, since local models are stochastically selected from a large model set, SSA can ensure that the adversarial examples do not overfit specific weaknesses of local source models. Extensive experiments on the ImageNet dataset and NeurIPS 2017 adversarial competition dataset show the effectiveness of SSA in improving the transferability of adversarial examples and reducing the memory consumption of the training process.

Characterizing the time course of decision-making in change detection.

We propose a novel modeling framework for characterizing the time course of change detection based on information held in visual short-term memory (VSTM). Specifically, we seek to answer whether change detection is better captured by a first-order integration model, in which information is pooled from each location, or a second-order integration model, in which each location is processed independently. We diagnose whether change detection across locations proceeds in serial or parallel and how processing is affected by the stopping rule (i.e., detecting any change vs. detecting all changes; Experiment 1) and how the efficiency of detection is affected by the number of changes in the display (Experiment 2). We find that although capacity is generally limited in both tasks, the architecture varies from parallel self-terminating in the OR task to serial self-terminating in the AND task. Our novel framework allows model comparisons across a large set of models ruling out several competing explanations of change detection. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

A comparative study on the performance of multiphase flow models against offshore field production data

In the oil industry, multiphase flow models are commonly used to simulate the flow from reservoir to the production unit. Flow simulations help in the production monitoring and optimization to support the decision-making process. Despite the need for accurate simulations to support production operations, multiphase flow models are mostly developed using laboratory data and rarely validated under field-scale conditions. To improve the understanding of multiphase flow models in field conditions, this work evaluates the performance of a large set of models for a dataset composed of two production units with 20 producing wells and 865 measured production points. The wells are divided and analyzed in two segments: the surface flowline and the wellbore. The evaluation compares the models' performances using various statistical parameters and trending charts. The evaluation analyzes the impacts of production parameters, well geometry, predicted flow pattern, and pressure drop components on the model performances. Overall, Ansari's model presented the best performance with 13.8% absolute average percent error for the flowline segment and 14.1% error for the wellbore segment. In addition, Gray can also generate good results for most but not all of the tested wellbores. From the production parameters evaluation, higher flow rate tests resulted in more stable outputs. Ansari and Beggs models showed a trend of increasing errors for wells with larger horizontal wellbore lengths. Also, Beggs model showed larger errors when the transition flow pattern was predicted. Finally, both Ansari and Beggs models showed larger overprediction errors for tests with higher pressure losses in the flowline segment, while Gray showed smaller errors for these cases. The conclusions can be used to select the best models given the production system's flow parameters, or be incorporated in systems to increase the accuracy.

In silico simulations reveal that RYR distribution affects the dynamics of calcium release in cardiac myocytes.

The dyads of cardiac myocytes contain ryanodine receptors (RYRs) that generate calcium sparks upon activation. To test how geometric factors of RYR distribution contribute to the formation of calcium sparks, which cannot be addressed experimentally, we performed in silico simulations on a large set of models of calcium release sites (CRSs). Our models covered the observed range of RYR number, density, and spatial arrangement. The calcium release function of CRSs was modeled by RYR openings, with an open probability dependent on concentrations of free Ca2+ and Mg2+ ions, in a rapidly buffered system, with a constant open RYR calcium current. We found that simulations of spontaneous sparks by repeatedly opening one of the RYRs in a CRS produced three different types of calcium release events (CREs) in any of the models. Transformation of simulated CREs into fluorescence signals yielded calcium sparks with characteristics close to the observed ones. CRE occurrence varied broadly with the spatial distribution of RYRs in the CRS but did not consistently correlate with RYR number, surface density, or calcium current. However, it correlated with RYR coupling strength, defined as the weighted product of RYR vicinity and calcium current, so that CRE characteristics of all models followed the same state-response function. This finding revealed the synergy between structure and function of CRSs in shaping dyad function. Lastly, rearrangements of RYRs simulating hypothetical experiments on splitting and compaction of a dyad revealed an increased propensity to generate spontaneous sparks and an overall increase in calcium release in smaller and more compact dyads, thus underlying the importance and physiological role of RYR arrangement in cardiac myocytes.

Climate model projections from the Scenario Model Intercomparison Project (ScenarioMIP) of CMIP6

Abstract. The Scenario Model Intercomparison Project (ScenarioMIP) defines and coordinates the main set of future climate projections, based on concentration-driven simulations, within the Coupled Model Intercomparison Project phase 6 (CMIP6). This paper presents a range of its outcomes by synthesizing results from the participating global coupled Earth system models. We limit our scope to the analysis of strictly geophysical outcomes: mainly global averages and spatial patterns of change for surface air temperature and precipitation. We also compare CMIP6 projections to CMIP5 results, especially for those scenarios that were designed to provide continuity across the CMIP phases, at the same time highlighting important differences in forcing composition, as well as in results. The range of future temperature and precipitation changes by the end of the century (2081–2100) encompassing the Tier 1 experiments based on the Shared Socioeconomic Pathway (SSP) scenarios (SSP1-2.6, SSP2-4.5, SSP3-7.0 and SSP5-8.5) and SSP1-1.9 spans a larger range of outcomes compared to CMIP5, due to higher warming (by close to 1.5 ∘C) reached at the upper end of the 5 %–95 % envelope of the highest scenario (SSP5-8.5). This is due to both the wider range of radiative forcing that the new scenarios cover and the higher climate sensitivities in some of the new models compared to their CMIP5 predecessors. Spatial patterns of change for temperature and precipitation averaged over models and scenarios have familiar features, and an analysis of their variations confirms model structural differences to be the dominant source of uncertainty. Models also differ with respect to the size and evolution of internal variability as measured by individual models' initial condition ensemble spreads, according to a set of initial condition ensemble simulations available under SSP3-7.0. These experiments suggest a tendency for internal variability to decrease along the course of the century in this scenario, a result that will benefit from further analysis over a larger set of models. Benefits of mitigation, all else being equal in terms of societal drivers, appear clearly when comparing scenarios developed under the same SSP but to which different degrees of mitigation have been applied. It is also found that a mild overshoot in temperature of a few decades around mid-century, as represented in SSP5-3.4OS, does not affect the end outcome of temperature and precipitation changes by 2100, which return to the same levels as those reached by the gradually increasing SSP4-3.4 (not erasing the possibility, however, that other aspects of the system may not be as easily reversible). Central estimates of the time at which the ensemble means of the different scenarios reach a given warming level might be biased by the inclusion of models that have shown faster warming in the historical period than the observed. Those estimates show all scenarios reaching 1.5 ∘C of warming compared to the 1850–1900 baseline in the second half of the current decade, with the time span between slow and fast warming covering between 20 and 27 years from present. The warming level of 2 ∘C of warming is reached as early as 2039 by the ensemble mean under SSP5-8.5 but as late as the mid-2060s under SSP1-2.6. The highest warming level considered (5 ∘C) is reached by the ensemble mean only under SSP5-8.5 and not until the mid-2090s.

Stellar Collapse Diversity and the Diffuse Supernova Neutrino Background

Abstract The diffuse cosmic supernova neutrino background (DSNB) is an observational target of the gadolinium-loaded Super-Kamiokande (SK) detector and the forthcoming JUNO and Hyper-Kamiokande detectors. Current predictions are hampered by our still incomplete understanding of the supernova (SN) explosion mechanism and of the neutron star (NS) equation of state and maximum mass. In our comprehensive study we revisit this problem on grounds of the landscapes of successful and failed SN explosions obtained by Sukhbold et al. and Ertl et al. with parameterized one-dimensional neutrino engines for large sets of single-star and helium-star progenitors, with the latter serving as a proxy for binary evolution effects. Besides considering engines of different strengths, leading to different fractions of failed SNe with black hole (BH) formation, we also vary the NS mass limit and the spectral shape of the neutrino emission and include contributions from poorly understood alternative NS formation channels, such as accretion-induced and merger-induced collapse events. Since the neutrino signals of our large model sets are approximate, we calibrate the associated degrees of freedom by using state-of-the-art simulations of proto-NS cooling. Our predictions are higher than other recent ones because of a large fraction of failed SNe with long delay to BH formation. Our best-guess model predicts a DSNB -flux of cm−2 s−1 with cm−2 s−1 in the favorable measurement interval of [10, 30] MeV and cm−2 s−1 with energies > 17.3 MeV, which is roughly a factor of two below the current SK limit. The uncertainty range is dominated by the still insufficiently constrained cosmic rate of stellar core-collapse events.

Modeling the stress-strain state of a circular plate subject to general corrosion according to various models

Quite a lot of works have been devoted to the problem of modeling the behavior of thin-walled structures exposed to an aggressive environment leading to corrosive wear of their surface. Researchers have proposed a fairly large set of models of corrosive wear, taking into account the influence of various factors on the kinetics of corrosion (time, material, temperature, the nature of the corrosive environment, the stress-strain state of the structure). Moreover, different authors often propose different models for the same conditions. In the article under consideration, a rather unique comparative study of three corrosion models proposed by different authors (Dolinsky V.M., Gutman E.M., Ovchinnikov I.G.) was carried out to simulate the behavior of the same circular plate subjected to the combined action of load and corrosion wear and tear. Moreover, the identification of the models, that is, the determination of the coefficients included in them, was carried out using the same experimental data. These models were then used to simulate the behavior of plates subject to corrosive wear under various loads. The results of numerical simulation were compared with experimental data obtained during testing of corrosive plates. Interestingly, in the models used, the effect of the stress state on the kinetics of corrosion was taken into account using different invariants of the stress state: the stress intensity in V.M. Dolinsky, medium voltage in the model of E.M. Gutman, and the specific energy in the model of I.G. Ovchinnikov. The analysis showed that the difference from the experiment when using the three models considered does not exceed 9.3%. The discrepancy between the results obtained using different models is also within the acceptable range, which suggests that all three models can be used to predict the behavior of plates under corrosive wear conditions. However, it is of interest to conduct research on the predictive capabilities of models on large forecast arms that go beyond the scope of experimental studies. At the same time, carrying out numerical experiments to simulate the behavior of complex structures in a stressed state and subject to corrosive wear, using several models that allow a good description of the experimental data and the most complete consideration of the operating conditions, makes it possible to obtain a more complete and versatile picture of what is happening in design processes, in comparison with the calculations performed according to one model, even if it describes the experimental data well.

Time-frequency forecast of the equity premium

Any time series can be decomposed into cyclical components fluctuating at different frequencies. Accordingly, in this paper, we propose a method to forecast the equity premium which exploits the frequency relationship between the equity premium and several predictor variables. We evaluate a large set of models and find that, by selecting the relevant frequencies for equity premium forecasting purposes, this method significantly improves in a statistical and economic way upon standard time series forecasting methods. This outperformance is robust regardless of the predictor used, the out-of-sample period considered, and the frequency of the data used.