Stochastic Dependence Research Articles

Формування податкової політики у країнах Європейського Союзу

The aim of the article is to generalize and systematize the provisions on the formation of tax policy in the EU countries. The dynamics of the indicator of the ratio of tax revenues to GDP in 27 EU countries during the period 2013–2022 is studied. It is found that the average value of this indicator for the EU-27 is 37.0%. A rather significant differentiation of the tax burden in the context of European countries is recorded, the lowest value observed in Ireland – 24.0%, and the maximum – in France – 47.4%. In general, a moderate level of tax burden is observed in the sample of Central European and Baltic countries. Using the method of correlation and regression analysis, it is found that for the EU countries in 2013–2022, with an increase in the tax burden by 1 percentage point, the real growth rate of the economy slows down by 0.215 percentage points, and the stochastic dependence is significant, but not too strong. In further research, it is important to pay attention to other factors influencing the growth rate of the real economy. Tax service is a key institution in the field of public revenue mobilization, in the EU-27 countries 62% of the total State revenues are collected by the tax authorities. The article pays attention to the study of the compositional tax structure both in the context of taxation of factors of production and individual taxes. Labor taxes account for the highest share in the tax structure in the EU Member States, followed by consumption taxes, and capital taxes are the least fiscally important. The determinant of the formation of the tax policy of European countries is competition in the international market to attract the main factors of production – labor and capital, which necessitates the creation of favorable conditions for their taxation. In the EU countries, there are unified taxation rules that apply to both minimum tax rates (VAT, excise duty) and control procedures, the principles of production and circulation of excisable goods, etc. In the EU member states, a standard and reduced VAT rate is used, the latter is a fiscal tool for maintaining social stability in society. The priorities of the tax authorities in the EU countries are the improvement of administration procedures, the expansion and modification of digital taxpayer services, the implementation of the institutional model of the State tax risk management and scenario modeling.

Sequence-to-Sequence Recurrent Graph Convolutional Networks for Traffic Estimation and Prediction Using Connected Probe Vehicle Data

Traffic estimation is imperative for conducting fundamental transportation engineering tasks such as transportation planning and traffic safety studies. Additionally, traffic prediction is vital for many data-driven intelligent transportation system applications. Most traffic estimation and prediction methods rely on infrastructure-based sensors to collect traffic parameters. However, infrastructure-based data collection can be costly and time consuming to set up and maintain. Additionally, the spatial distribution of the collected traffic data is limited by the location of the deployed hardware sensors. Probe vehicle data can be used to overcome these limitations. Traffic modeling for estimation and prediction is a complex task due to the stochastic nonlinear spatiotemporal dependencies exhibited by traffic parameters. In this paper, a deep learning-based sequence-to-sequence architecture called Seq2seq GCN-LSTM was proposed to estimate and predict network-wide traffic volume and speed. The proposed method utilizes short-term historical traffic data collected from a low-penetration rate probe vehicle fleet to estimate and predict traffic parameters up to 60-minutes ahead. The method utilizes Graph Convolutional Networks for spatial dependency extraction and Long Short-Term Memory networks to model temporal dependencies. The proposed method generated superior traffic results compared to the baseline models. Additionally, the model demonstrated robustness against perturbations caused by the low rate. Furthermore, the probe vehicle penetration rate was varied to test its effect on the proposed method’s modeling capability. The model was able to maintain traffic volume and speed estimation and prediction performance within a reasonable margin of error using penetration rates as low as 1.5% and 0.5%, respectively.

��²�� �� ������� ����� ���ֲ ������� � ��в�� �Ϊ����� �����

An urgent issue is to conduct systematic studies of the education role in the Ukrainian labor market formation and functioning. The study�s purpose is to identify the existing prerequisites and determine directions for the educational sector development as one of the Ukrainian labor market defining subjects. For this, several complex tasks of a comparative analysis of Ukraine�s official statistical data in peace (2021) and war (2022) years are being solved. The general methodological orientation is to identify relative proportions, taking into account the martial law impact on the labor market and the statistical information reliability. Macroeconomic indicators of the Ukraine labor market are studied by the methods of structural and diversified intersectional comparison and statistical analysis of stochastic dependencies under uncertain conditions. Theoretical generalizations of changes in macroeconomic indicators and the interpretation for the future are based on the results of the structural-dynamic analysis. Under martial law, there was a reduction in supply and demand by almost 2 times against the backdrop of a significant increase in market imbalance. According to the results of the statistical data processing, the direction and form of education activation as a catalyst for the recovery and development of the economy of Ukraine are proposed. The initial direction in the current Ukraine labor market is to use the education potential to eliminate the jobs lacking. Providing knowledge to the unemployed to fill existing vacancies remains the education prerogative. The form of implementation of these directions is the diversification of education on the basis of public-private partnerships. The public-private institutional field appropriate adjustment in the form of training and qualification complexes (primarily in the vocational education area) is a step toward the public-private partnerships introduction in other economic sectors. Further research should be focused on finding ways to activate education as the labor market subject of forming quantitative and qualitative parameters. For example, in relation to the required and available qualifications of the unemployed, working conditions, and the level of remuneration.

The Philosophy of Copula Modeling: A Conversation with ChatGPT

In the form of a scholarly exchange with ChatGPT, we cover fundamentals of modeling stochastic dependence with copulas. The conversation is aimed at a broad audience and provides a light introduction to the topic of copula modeling, a field of potential relevance in all areas where more than one random variable appears in the modeling process. Topics covered include the definition, Sklar’s theorem, the invariance principle, pseudo-observations, tail dependence and stochastic representations. The conversation also shows to what degree it can be useful (or not) to learn about such concepts by interacting with the current version of a chatbot.

Data-driven spatiotemporal modeling for structural dynamics on irregular domains by stochastic dependency neural estimation

Numerical simulations for spatiotemporal processes involving material, geometrical and contact nonlinearities might be computationally prohibitive for many-evaluation applications. Considering the strong nonlinearity of dynamic processes, the prediction of high-dimensional spatiotemporal responses remains challenging for most surrogate-based models. To address such issues, a Stochastic Dependency Neural Estimator (SDNE) is proposed to construct the mapping from the input parameter domain to the high-dimensional spatiotemporal response domain, revealing the underlying relationship between the high-dimensional regression estimation and manifold distribution hypothesis. Moreover, Graph Neural Networks (GNN) and Temporal Convolutional Neural Networks (TCN) are integrated to capture nonlinear features from irregular geometries and time series simultaneously. Two high-speed impact cases involving large deformation and highly nonlinear transients are investigated to validate the proposed method. The proposed SDNE achieves highly accurate prediction compared with Finite Element (FE) evaluations. To further validate the practical application of the proposed method, multi-objective crashworthiness optimizations are performed with the proposed SDNE. The results show that the proposed SDNE provides reliable predictions for spatiotemporal dynamics

Degradation modeling and remaining useful life prediction for a multi-component system with stochastic dependence

Stochastic dependence between components within a system implies that the degradation state of a component influences the lifetime distribution of the other components. The potential failure risks of these components will be underestimated if these effects are ignored. Therefore, the objective of this study was to investigate the effects of the stochastic dependence between components on the degradation process and remaining useful life (RUL) of a system. Firstly, a degradation model integrating the effects of stochastic dependence between components was formulated. Then, the probability density function of the RUL was derived for multi-component systems with different structures. Finally, the dependent degradation state and unknown parameters of the model were estimated simultaneously and recursively using Kalman filtering in conjunction with the expectation maximization algorithm. The superiority of the presented method was confirmed by considering a numerical example and performing case studies of an aircraft turbine engine and a gearbox system.

Stochastic Factor Analysis of Occupational Injuries at Railway Transport Enterprises

Introduction. Railway enterprises are characterized by an increased labor risk, the professional activity of personnel is associated with the risk of injury. Occupational safety measures are regularly carried out in the industry. At the same time, accidents are known to occur in Russian Railways JSC (Russian Railways). Risk reduction requires a theoretical study of the problem, as well as the study of applied solutions, which determines the relevance and significance of the proposed study. The purpose of the work is to determine the resulting indicators on which the dynamics of industrial injuries in Russian Railways depends.Materials and Methods. To identify the causes of injury to railway transport workers of the Russian Railways holding, the authors analyzed statistics for 2007–2021. We systematized and ranked 17 types of accidents, which are associated with accidents resulting in injuries. Using the Pareto chart, they were differentiated as resultant and non-resultant. We identified 7 resultants. We performed a stochastic analysis and established the relationship of each factor with the total number of accidents. The correlation coefficients were calculated.Results. Preliminary calculations were presented in the form of a table to obtain the coefficient of stochastic dependence of industrial injuries on the number of injured employees of Russian Railways in road accidents. The value of the stochastic relations coefficient was calculated and adjusted. The average error was determined — the difference between the general injury rate and the injuries in road accidents. The results of these calculations were compared with the data of the Student's t-distribution quantile table for confidence probability. Similarly, the degree of stochastic relationships for the other resulting types of incidents was calculated and summarized in a table. The result was visualized in a diagram.Discussion and Conclusion. The results of the conducted surveys allow us to assert that the dynamics of industrial injuries in Russian Railways JSC is determined mainly by incidents in road accidents. It is necessary to investigate their causes and develop measures aimed at improving occupational safety.

Distribution-free estimation of individual parameter logit (IPL) models using combined evolutionary and optimization algorithms

When estimating random coefficients models from choice data, decisions relating to the multivariate density function assumed to describe preference heterogeneity across the population raise questions about stochastic (in)dependence between preference dimensions, uni- vs. multi-modality, potential point masses, bounds and/or constraints on support regions, among other concerns. Parametric representations of population distributions have generally implied uncomfortable compromises to achieve estimation tractability. It would seem preferable to sidestep such issues by estimating individual preferences in a distribution-free manner, but this freedom of form implies a large number of parameters since we lose the parsimony enabled by parametric densities and must deal directly with estimation of individual decision maker preferences. I propose a hybrid distribution-free estimator for individual parameter logit models that uses a genetic algorithm as first stage, the solution from which becomes a starting point for a gradient-based search to obtain the final posterior maximum likelihood estimates of individual preferences. This estimator is described in detail, its parameter recovery capability is tested with Monte Carlo data generation simulations, and a case study is developed in some detail to illustrate its use in policy analysis. The estimator can be applied to both stated and revealed preference data, requiring only that sufficient choice replications be available for individual observation units consistent with extant estimation methods. Computational experience shows the estimator to require CPU times comparable to extant simulation-based estimation methods, meaning that its use is practical for the exploration of the parameter space through multiple trials.

Aggregated Net-load Forecasting using Markov-Chain Monte-Carlo Regression and C-vine copula

Net-load is the difference between total load and renewable generation and acts as an effective system load to which dispatchable generators are scheduled and system flexibility requirements are estimated. This necessitates accurate net-load forecasts for optimum scheduling and flexibility requirement estimations. Net-Load Forecasting (NLF) got only little attention in existing literature even though it is essential for optimal generation scheduling and power system flexibility requirement estimations. Time series and machine learning models have been used for NLF in recent years, however, there is vast scope existing for improving accuracy. Probabilistic forecasting models are widely used for various forecasting problems such as load, wind, and solar generation forecasting as those models show improved forecasting accuracy. In this context, this paper proposes a novel probabilistic aggregated very short-term NLF model based on Markov Chain Monte Carlo (MCMC) Regression. MCMC uses data augmentation, where unobserved variables are simulated from their posterior distribution and this makes MCMC approach suitable for regression. Further, MCMC forecasts are improved by incorporating C-vine copula-based Joint Probability Distribution (JPD) of expected load, wind, and solar generation forecasting errors. The C-vine copula is suitable for such multi-variable JPD estimation due to its enormous flexibility in stochastic multivariate dependence modelling compared to standard elliptical and Archimedean copulas. Results show that the proposed NLF model outperforms reference models and can produce accurate net-load forecasts.

A new family of modified Gaussian copulas for market consistent valuation of government guarantees

This paper deals with a copula-based stochastic dependence problem in the context of financial risks. We discuss the financial framework for assessing the theoretical up-front value of government guarantees on bank liabilities. EU States widely use these contracts to improve the financial system’s stability and manage the banking sector in crisis situations; in Italy, they have also been used to address the consequences of the Covid-19 emergency. From a market viewpoint, we deal with a defaultable guarantee contract where the State-guarantor and the bank-borrower are both subject to default risk, and their risks are interconnected. We show that the classical Gaussian copula is not satisfactory for modeling the dependence among the considered risks. Indeed, using the benchmark market model for credit risk portfolio management, we highlight some contradictory results observed for the up-front values of the guarantee when the default intensity of the guarantor is smaller than that of the borrower. Then, we introduce a new family of modified Gaussian copulas that overcomes the limitations of the standard approach, allowing to determine realistic results in terms of the guarantees “mark-to-model” value when the benchmark market model does not work. Numerical simulations validate the theoretical proposal.

Problem of Modeling and Control of Multi-Dimensional Timeless Processes with Dependent Output Variables

In this paper, we consider the problems of modeling and controlling multidimensional inertialess systems with delay under nonparametric uncertainty. We are talking about the multidimensional processes which are in conditions where the form of parametric equations for various channels of the object is absent due to lack of a priori information. The main emphasis is placed on the case when the components of the vector of output variables are randomly connected in an unforeseen way. In the case of a stochastic dependence of the output variables, the mathematical description of the object is reduced to a system of implicit equations, the form of which is unknown. Therefore, the main task of modeling is to find the predicted values of output variables from known input variables. When managing a multidimensional object, an important feature is the definition of setting actions. The main thing here is that the setting influences of the control system should not be chosen arbitrarily from the corresponding areas, but depending on the definition of the previous ones. Nonparametric identification and control algorithms for multidimensional systems are proposed. Computational experiments that show the efficiency of using the proposed nonparametric identification and control algorithms are presented.

Inspection and maintenance optimization for heterogeneity units in redundant structure with Non-dominated Sorting Genetic Algorithm III

Redundant structure has been widely deployed to improve system reliability, as when one unit fails, the system can continue to function by using another one. Most existing studies rely on the similar assumption that the heterogeneous units are subject to periodic inspections and identical in terms of their aging situations and the numbers of resisted shocks. In practice, it is often adequate to trigger a unit individually in the event of a single shock, which intensifies the degradation of that unit, accordingly, requiring a sooner inspection to ensure its safety. In this study, the stochastic dependency among units is addressed firstly by introducing a novel activation sequence. Secondly, an adaptive system-level inspection policy is proposed by prioritizing the unit with a worse state. Finally, we take advantage of Monte Carlo methods to simulate the whole process and estimate two objectives, referring to the average system unavailability and maintenance cost, in a designed service time. It is found that the two objectives are contradictory through numerical examples. The Non-dominated Sorting Genetic Algorithm III (NSGA-III) algorithm, therefore, has been employed to find the optimal solutions in system unavailability and cost, which provide clues for practitioners in decision-making.

Physics-Informed Data-Driven Prediction of 2D Normal Strain Field in Concrete Structures.

Concrete exhibits time-dependent long-term behavior driven by creep and shrinkage. These rheological effects are difficult to predict due to their stochastic nature and dependence on loading history. Existing empirical models used to predict rheological effects are fitted to databases composed largely of laboratory tests of limited time span and that do not capture differential rheological effects. A numerical model is typically required for application of empirical constitutive models to real structures. Notwithstanding this, the optimal parameters for the laboratory databases are not necessarily ideal for a specific structure. Data-driven approaches using structural health monitoring data have shown promise towards accurate prediction of long-term time-dependent behavior in concrete structures, but current approaches require different model parameters for each sensor and do not leverage geometry and loading. In this work, a physics-informed data-driven approach for long-term prediction of 2D normal strain field in prestressed concrete structures is introduced. The method employs a simplified analytical model of the structure, a data-driven model for prediction of the temperature field, and embedding of neural networks into rheological time-functions. In contrast to previous approaches, the model is trained on multiple sensors at once and enables the estimation of the strain evolution at any point of interest in the longitudinal section of the structure, capturing differential rheological effects.

Stochastic fixed-point equation and local dependence measure

We study solutions to the stochastic fixed-point equation X=dAX+B where the coefficients A and B are nonnegative random variables. We introduce the “local dependence measure” (LDM) and its Legendre-type transform to analyze the left tail behavior of the distribution of X. We discuss the relationship of LDM with earlier results on the stochastic fixed-point equation and we apply LDM to prove a theorem on a Fleming–Viot-type process.

Chance-constrained co-optimization for day-ahead generation and reserve scheduling of cascade hydropower–variable renewable energy hybrid systems

The rapid development of variable renewable energy (VRE), such as wind and solar energy, has stimulated the complementary operation of VRE with flexible cascade hydropower stations in China. Due to the uncertainties of VRE power generation and complex constraints of cascade hydropower stations, formulating reliable day-ahead generation and reserve scheduling plans is a real challenge in the actual operation stage of the cascade hydropower–VRE hybrid systems (CHVHS). In this paper, we propose a tractable chance-constrained co-optimization model for day-ahead generation and reserve scheduling of a CHVHES. First, compared with existing models, the relationship between the hydropower reserve capacities and water-electricity conversion efficiency is finely modeled. Accordingly, the economic allocation of total VRE reserve requirements among cascade hydropower stations is for the first time considered in our proposed model. This can save more water resources for cascade hydropower stations when compensating for VRE, further improving hydro–VRE complementary profits. Second, we propose a solution approach for chance constraints by incorporating the nonparametric probabilistic forecasting of VRE power based on quantile regression into the chance-constrained model, ensuring that the stochastic dependence between VRE power output and its point forecast can be effectively captured. Importantly, this solution approach does not require prior knowledge or any probability distribution assumptions of VRE power and does not introduce any additional computational burden. With the help of three-dimensional interpolation technology for nonlinear constraints, the proposed scheduling model is finally cast as a mixed-integer linear programming model that is computationally tractable. Numerical tests implemented on a real CHVHES located in Southwest China verify the effectiveness and advantages of the proposed methods.

A spatial model to jointly analyze self-reported survey data of COVID-19 symptoms and official COVID-19 incidence data.

This work presents a joint spatial modeling framework to improve estimation of the spatial distribution of the latent COVID‐19 incidence in Belgium, based on test‐confirmed COVID‐19 cases and crowd‐sourced symptoms data as reported in a large‐scale online survey. Correction is envisioned for stochastic dependence between the survey's response rate and spatial COVID‐19 incidence, commonly known as preferential sampling, but not found significant. Results show that an online survey can provide valuable auxiliary data to optimize spatial COVID‐19 incidence estimation based on confirmed cases in situations with limited testing capacity. Furthermore, it is shown that an online survey on COVID‐19 symptoms with a sufficiently large sample size per spatial entity is capable of pinpointing the same locations that appear as test‐confirmed clusters, approximately 1 week earlier. We conclude that a large‐scale online study provides an inexpensive and flexible method to collect timely information of an epidemic during its early phase, which can be used by policy makers in an early phase of an epidemic and in conjunction with other monitoring systems.

Assessment of the permeability and pore size of quartz-based consolidated sedimentary rocks by dielectric spectroscopy in the frequency range of 1 kHz - 500 MHz

The spectral-induced polarization (SIP) response of porous materials at frequencies below 1 kHz can be used to determine the pore size and rock permeability. Although the results of certain studies are encouraging, in recent years it has been argued that there cannot be a universally applicable approach to predict permeability using the SIP response. This article proposes a broadband dielectric spectroscopy method in the frequency range from 100 Hz to 500 MHz for assessing the permeability and average pore size. An original experimental setup has been developed that makes it possible to measure the dielectric spectrum of quartz-based consolidated sedimentary samples of porous rocks in this frequency range. The rock samples with permeability from 0.17 to 206 mD and an average pore size from 6.3 to 21.1 μm have been studied during the experiments. The dielectric spectra of four sandstone samples and two siltstone samples moistened with distilled water and a NaCl solution with a concentration of 1.5 and 4 g/L were measured. The volume fraction of the solution in the samples varied from complete saturation to a moisture of 0.04–0.05 m3/m3. The dielectric spectra were modeled by three relaxation processes, the parameters of which were determined using the Debye and Cole-Cole formulas. The strength of two low-frequency processes with the relaxation times of about 1.2 ms and 0.15 ms are well correlated with the low-frequency conductivity of the samples; however, no stochastic dependence on the pore size and permeability was found. The parameters of the high-frequency relaxation process with the relaxation times of tens and hundreds of nanoseconds depend both on the solution ratio in the sample and its concentration, as well as on the pore size and permeability. The strength of this process is better related to the pore size and permeability when moistened with distilled water, and the relaxation time correlates best with them when using the solution with a concentration of 4 g/L. No significant differences in the process parameters in the samples of sandstone and siltstone were found.

Optimizing System Reliability in Additive Manufacturing Using Physics-Informed Machine Learning

Fused filament fabrication (FFF), an additive manufacturing process, is an emerging technology with issues in the uncertainty of mechanical properties and quality of printed parts. The consideration of all main and interaction effects when changing print parameters is not efficiently feasible, due to existing stochastic dependencies. To address this issue, a machine learning method is developed to increase reliability by optimizing input parameters and predicting system responses. A structure of artificial neural networks (ANN) is proposed that predicts a system response based on input parameters and observations of the system and similar systems. In this way, significant input parameters for a reliable system can be determined. The ANN structure is part of physics-informed machine learning and is pretrained with domain knowledge (DK) to require fewer observations for full training. This includes theoretical knowledge of idealized systems and measured data. New predictions for a system response can be made without retraining but by using further observations from the predicted system. Therefore, the predictions are available in real time, which is a precondition for the use in industrial environments. Finally, the application of the developed method to print bed adhesion in FFF and the increase in system reliability are discussed and evaluated.

Dynamic control for construction project scheduling on-the-run

Construction project management requires dynamic mitigation control to ensure a project's timely completion. Current mitigation approaches are usually performed by an iterative Monte Carlo (MC) analysis which does not reflect (1) the project manager's goal-oriented behavior, (2) contractual project completion performance schemes, and (3) stochastic dependence between construction activities. Therefore, the development statement within this paper is to design a method and implementation tool that properly dissolves all of the aforementioned shortcomings ensuring the project's completion date by finding the most effective and efficient mitigation strategy. For this purpose, the Mitigation Controller (MitC) has been developed using an integrative approach of nonlinear stochastic optimization techniques and probabilistic Monte Carlo analysis. MitC's applicability is demonstrated using a recent Dutch large infrastructure construction project showing its added value for dynamic control on-the-run. It is shown that the MitC is a state-of-the-art decision support tool that a-priori automates and optimizes the search for the best set of mitigation strategies on-the-run rather than a-posteriori evaluating the potentially sub-optimal and over-designed mitigation strategies (as commonly done with modern software such as Primavera P6).

A predictive maintenance model for [formula omitted]-out-of-[formula omitted]:F continuously deteriorating systems subject to stochastic and economic dependencies

New concepts of Industry 4.0 make industrial systems more and more complex due to the strong dependencies among a large number of stochastically deteriorating units. Ignoring such dependencies in maintenance modeling usually leads to high maintenance costs. With this spirit, we develop in the present paper an innovative predictive maintenance model for k-out-of-n:F continuously deteriorating systems considering both the stochastic dependency due to the common impact of working environment and the economic dependency issued by the set-up cost sharing. Compared with related works, the originality of our model is fourfold. Firstly, we adopt a multivariate Gamma subordinator to describe the dependent degradation behavior among units in the context of maintenance modeling. Secondly, we use probabilistic measures of the remaining useful life at unit and system levels as maintenance decision indices. Thirdly, we elaborate a proactive maintenance policy allowing the consideration of both the stochastic and economic dependencies in maintenance decision-making. Finally, we evaluate analytically the long-run cost rate of the maintenance policy thanks to the semi-regenerative theory. Various comparative studies with benchmarks under different configurations of system characteristics and maintenance costs confirm the high performances of the developed maintenance model, and emphasize the value of introducing multiple kinds of dependencies in maintenance modeling.